# A Disturbing Chart, from Bruno Ombreux

December 2, 2014 | 3 Comments

A disturbing chart: "This is Probably the Second Worst Time in History to Own Stocks"

## Bill Rafter writes:

The trouble with the chart is that the regression fit was done cumulatively, resulting in older data being subject to look-ahead bias. Thus only the current values are useful, and one wonders exactly how useful. As Steve has commented, the way to foil that is to use a moving regression fit in which the values are static over time, always taking the last point in the fit. Thus all data, past and current are relevant and can then be used in statistical studies.

The question that then comes up is which lookback period do you use. Wherever possible all lookback periods should be adaptive, the question then being to what input. In shorter term price data the market will tell you the relevant lookback period. I have never tried determining lookbacks for longer term data because (a) I don't expect to live long enough to take advantage of it, and (b) too many things can happen in the short run to screw up a good plan. Most people don't marry someone in their 20s based on the supposition that (s)he will look good in their 70s.

I also question the use of any equity or debt data prior to 1972. If you don't know why, ask Stefan. **That's one of the great things about the list; there are sources for just about everything.

Several moving functions you should consider:

Moving linear (i.e., regression) fits and their slopes.

Moving parabolic fits and their slopes. Since most economic and price data are parabolic, this is the better of the two. There is also something to be gained in the difference between a parabolic fit and a linear fit. Fitting parabolas is quite tricky, and it took us a while to code it. If you try to do so and want a check on your efforts, try fitting a parabola to a straight line. If the result is ludicrous, try a different method.

Moving correlations are particularly interesting between markets that might be alternatives to one another. Moving correlations between stocks and bonds (levels to levels) are something we have used for years and continue to do so. I thank Gibbons for his comment that Colby & Myers recommended them, as I had not been aware of that. (I'm not a fan of C&M.)

## Gyve Bones responds:

Colby and Myers didn't recommend the linear regression study per se… the empirical analysis simply showed that study to perform best with a fixed loopback parameter over NYSE index returns data over a long period of time compared to other trend following signal generators. This book was an early attempt to quantify different approaches to see how they performed trying as best as can be done to compare apples to apples. In the mid-to-late 80s, it was the best thing that had been done like that since Dunn & Hargitt's study using punch card futures data in the late 1960s (which found that the Donchian Four Week system was best, the system which launched a thousand CTA, including the Dennis Turtles and their spawn.) Another similar study was done in the 90s by Jack Schwager and another fellow whose name escapes me at the moment which was well done.

A question: when was the regression line fit? Today? 20 years ago? 50 years ago? The slope will change based on your starting and end points. How overbought or sold is a function of this. A more careful analysis would either apply this same "method" every year with a set of rules (i.e sell above x% overbought) or would do the same thing on a rolling window basis. It's an interesting chart nonetheless and gives one pause, but I would suggest it lacks a certain amount of rigor.

## Gibbons Burke writes:

It seems to me that this is a flawed chart to look at historically to make rules from because the trend line drawn into the past contains information about the future. The line is drawn using the linear regression of the entire data set so, for example, the line segment covering 1998-1999 "knows" about what happened in 2014. Very deceptive and misleading to make a rule based on the relationship of the data to the trend line.

The disturbing chart is a case study of why charting is so misleading because of the regression bias and also at the variance of a sum is the sum of the variances.

## Steve Ellison says:

Here is the way to solve the problem of the regression line incorporating future data. Attached is a graph of a "moving regression", as Dr. Rafter calls it. For each date, the red point is the last point of a 30-year regression of the S&P 500 as of that date (the graph is from 2010).

# Germany and Coal, from Bruno Ombreux

Germany has absolutely no problem with power supply. In the past few years it has turned from net importer of power to next exporter. On the supply side, renewables have had a huge impact. There are some sunny and windy week-ends when the country is running 100% on renewables. On the demand side, their consumption has been decreasing a couple percents a year. This trend of consumption decrease is expected to continue, because of big energy efficiency plans.

The only problem in Germany is one problem of imbalance. They have excess power in the North and not enough in the South. As a result, their power is flowing from North to South, but thru the lowest impedance path, which is neighboring countries. In particular, they are messing up the Czech Republic grid. There was a bit of a diplomatic row. This is being solved however, with the planned installation on phase shifters at the Czech border, and the planned connection of the German North to Norway (to use Norway's hydro has a means of storage of their excess power).

I strongly doubt any new nuclear plant will be built in any Western European country in the near future. Nuclear is very "has been".

# Thermodynamics and the Market, from Bruno Ombreux

Does Prigogine's principle have any predictive market implications?

Well if you move from thermodynamics to information theory entropy, and consider the information content of market prices, then there are two clear analogies:

1. There should be local, transient edges (Prigogine, market prices self-organizing to minimize the rate of information loss).

2. Those edges are decaying (Second Principle, "Changing cycles").

# The Trouble With Exponential Growth, from Jim Sogi

April 23, 2014 | 1 Comment

People generally cannot understand or have difficulty understanding or picturing exponential growth. There is a story about the inventor of chess when the King condescendingly asked him what he wanted for a reward, the inventor replied that he would like the number of grains of rice, which when starting with only one grain, doubled in amount for each square on the chess board. The King laughed and thought he got off so easy. This same inability to picture compounding growth interferes with a long compounding hold of financial assets.
The same lack of ability to see compounding growth applies to study of past growth. People understand linear growth more easily.

## Vince Fulco writes:

The inability to see compounding growth interferes with the study of nearly everything I might add. We tend to think that evolution is responsible for much if not all of the world that we see, a function of random mutations that have a selective advantage. Consider, after all, the universe is about 18 billion years old. In seconds, this is 60 x 60 x 24 x 365.2425 x 18 x 1,000,000,000 = 5.68e17.

"Random mutations with a selective advantage." Yes indeed. A thousand monkeys wailing away on typewriters will eventually happen upon Hamlet.

So let's examine that (it provides an insight into the astounding character of exponential growth James writes about here). Only considering the line "To be or not to be," which comprise 18 characters. We will consider the space a character, and make no distinction between upper case and lower case, require no punctuation, but rather a keyboard for the little apes having only 27 keys. The probability of any one monkey typing only this line is 27 ^ 18 = 5.81e25

To put the relative differences into perspective, if I could take a thousand monkeys, 18 billion years ago, and permit them 5,000 keystrokes per second, we would have about an even money bet that, without regard to case or punctuation they might, with a probability of about .5, come up with something like "to be or not to be" and much less all of Hamlet.

Natural selection, whereas I do not contest its existence, does not explain a whole lot as clearly not enough time has elapsed since the big bang.

I'm wrestling constantly these days with allocation structures based on similar matters, where a copula of discrete outcomes (say, the copula of rolling a pair of dice) posses 21 possible, distinct outcomes such that the branch out across elapsed consecutive trials gets unfathomably large quite rapidly. Even with parallel processing (more monkeys, more typewriters) the problem reduces, but the scale remains too enormous still as a result of astounding nature exponential growth.

## Henry Gifford writes:

The relatively new field of epigenetics has some very interesting answers to the astute observation that random mutation alone would have taken too long.

## Stefan Jovanovich writes:

So, one starts with a "known" fact that is not a fact at all — monkeys sitting at a typewriter writing Hamlet. Then, one proves mathematically that the fact is not probable; and that, in turn, raises a question about the validity of the current best hypothesis for explaining the organic world around us — namely, that through natural competition fortunate mutations win.

It seems to me that we are all saddled with two very stupid terms and our minds wear them like blinders - Marx's word for the results of enterprise ("capitalism") and the Darwin's unfortunate choice of the word "selection" for his title. "Evolution", especially in the eyes of its admirers who try to turn its theory into a fact, somehow takes on the certainty of religious moral authority; it "explains" everything just as Marxism does. No, it doesn't; but Darwin's theory does withstand the Shakespeare test. For one thing, there is no evidence that any monkey of sense would go near a typewriter.

If Marx fudged the data wherever possible to make history say what did not, in fact, happen, Darwin did the very opposite. He thought there could never be enough data to "prove" his hypothesis; but he did take heart that there was, at least not yet, no data that proved it wrong. He does seem to have realized that he had been proven wrong in his choice for the title for his first edition. But he deserves a pass for that.

In choosing On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life, he was not endorsing the Southern way of life or Dickens, Carlisle, Kingsley and Ruskin's defense of Governor Eyre. Hardly. Darwin was one of the very few people who had the courage to speak up, along with Huxley, John Bright, and John Stuart Mill. In using the words "race" and "species" Darwin was using the biologist's definition - could male and females produce offspring. In the rare times when he was asked about human "races" Darwin was genuinely bewildered by the question; there was, in his view, demonstrably only one human species/race.

That leaves the point HG notes. Useful mutations may have had their randomness accelerated.

"Timescales of Genetic and Epigenetic Inheritance":

"According to classical evolutionary theory, phenotypic variation originates from random mutations that are independent of selective pressure. However, recent findings suggest that organisms have evolved mechanisms to influence the timing or genomic location of heritable variability. Hypervariable contingency loci and epigenetic switches increase the variability of specific phenotypes; error-prone DNA replicases produce bursts of variability in times of stress. Interestingly, these mechanisms seem to tune the variability of a given phenotype to match the variability of the acting selective pressure. Although these observations do not undermine Darwin's theory, they suggest that selection and variability are less independent than once thought."

## Jeff Watson writes:

We should revisit the Second Law of Thermodynamics, and how some scientists speculate that it enables the formation of life itself. There is some very good peer reviewed literature regarding the second law and life.

Anyways, there are some very interesting challenges to Stanley Miller's glass jar filled with water, methane, ammonia, and hydrogen.(I believe he got his PhD from the same place as the Chair). In Miller's experiment, he blasted it with an electric arc for a long time, and out of this primordial soup arose half the amino acids required for life.

Fifty years later and biochemists and physicists today are on the verge of creating an artificial organism that meets all the criteria of life.

In evolution it's the survival of the fittest, so how come so many species still exist?

## Gary Rogan writes:

There are multiple niches in the environment, so species specialize.  The famous Darwin finches were shown to adapt their overall size and beak size to different sub-environment both geographically and on the same island.  It's the same reason why Toyota doesn't operate supermarkets or Procter and Gamble doesn't own any airlines.  It's easier with conglomerates: at least they can get specialized departments, but imagine a lion competing with heat-loving microorganisms in under-water vulcanoes.

Now the jumps: just because not all species that died out left any identifiable remains, doesn't mean they didn't exist.  The absence of evidence is not evidence of absence.

## Bruno Ombreux writes:

Why is it hard to believe that matter will organize itself into a more
complex form when a very high temperature source of energy is in the
sky?

Matter will organize itself when it is in an open system subjected to an energy gradient. See Prigogine's principle.

# Putin Nominated, from Bruno Ombreux

What took so long? I do not know. The Putin nomination makes some sense. If you look at it objectively, they gave the prize to Obama, when Obama was barely elected president and had done nothing for World peace.

Putin has done a lot for world peace.

Firstly, he is keeping a tight leash on Russian hardliners that want nothing more than a return to the good old USSR (the article from this month's print version of Bloomberg Magazine about Sechin missed the point by so many yards that I will not even begin talking about it).

Secondly, he averted a war in Syria.

Thirdly, he is defending the right of people to self-determination in the Crimean part of former Ukraine.

Very objectively, he saved more lives than Mother Theresa.

# Trends, from Jim Sogi

Seems like the market has been rather trendy lately. Of course now that I've realized it its probably near the end of the trend. But that's the same thing I though at the beginning of the trend.

Mean reversion systems have difficulty in a trendy market, and simple TA things work well for trends if you're lucky.

## Rocky Humbert writes:

Mr. Sogi writes: "Mean reversion systems have difficulty in a trendy market, and simple TA things work well for trends if you're lucky."

I suggest that Mr. Sogi should have written: "Simple TA things have difficulty in a choppy market, and mean reversion systems work well if you're lucky."

Every single profitable trade requires a trend!

If you buy at 9:30am at a price of 100 and sell at 9:31 at a price of 100.25, there was a one minute trend. Call it whatever you want. But if you have two points connected by a line, that line is a trend.

The carpenter ants that live in my yard don't know that my neighbor has much better foraging.

## Steve Ellison writes:

As I understand the premise of trend following, it is allegedly good to identify the trend in place before placing one's trade and enter the market on the side of that trend. To say every profitable trade requires a trend seems a tautology to me and not useful since the statement refers to the trend that occurs after entry and hence cannot be known at the time of entry.

Rocky,

This is a semantic debate. It all depends how you define a trend. "Point A to point B" is a "line", not necessarily a "trend". There are actually formal definitions for "deterministic trends" and "stochastic trends". There are also statistical tests to check the presence of those trends.

Mean-reversion: you can make money in a market going from "point A to point A" instead of "point A to point B".

## anonymous writes:

Having spent a number of years in the trend-follower business, I can confirm that trend-following, as practised by some rather large CTAs, means betting on markets where models suggest the continuation of a move. So if the price went up from A to B, a trend follower would make bets where the move from B to C is in the same direction, whereas a mean-reverting player will try trade instruments that he believes will move back towards A.

Over the years, I have given much thought to the workings of the whole trend-following business, and its role in the market ecosystem. The Chairman's various critiques of the style are all valid, and worth heeding. Yet, properly understood, I believe trend-following remains a valid approach to trading. i.e., it is a trading style that exposes you to risk factors for which the market is willing to pay you.

A wise man once said, "There ain't no point in beating a dead horse. But there ain't no harm in it either."

We've all had this trend following discussion ad nauseum in the past, and the chair's pathological aversion to trend following is well known. So to avoid re-opening old wounds, I will re-offer the single most plausible and economically rational reason why trend-following can work and has worked. (That is, I'm not saying anything about whether it still works or will work in the future.)

In order to move a price, the market requires new information. And this new information takes time to disseminate among market participants. And during this period of dissemination and acceptance of a new perception, prices will appear to trend. If you are the first person to acquire and understand this new information, you are said to have a variant perception. If you are the second or third person to realize that there is new information, you are called a trend follower. And if you instinctively fade this perception as it disseminates through the market, you are either called a contrarian or Anatoly. Strictly speaking, a true contrarian, like a stopped clock, is right twice a day. And while this new information is disseminating through the market, there are obviously many opportunitities to profit.

Ultimately, however, a trend-follower is economically equivalent to a person who buys synthetic options or volatility. And a mean-revision trader is economically equivalent to a person who sells synthetic options or volatility. Transaction costs notwithstanding, unless one has superior information, there is no apriori reason to believe that selling synthetic options should, over a career, be more profitable than buying synthetic options. However, the equity profile of an options seller is that of many small profits and a few big losses. Whereas the equity profile of an options buyer is that of many small losses with a few big gains.

# The Oil Market, from Gordon Haave

December 14, 2012 | 1 Comment

Since I married an Albanian, I have become an olive fanatic, and I have to say the upcoming Californian olive industry is going to be a real challenge to Greece. Californian olives are fantastic, and presumably harvested much more efficiently than the Greek ones.

## Bruno Ombreux writes:

I've read Spain produces about four times as much olive oil as Greece, so if there is a threat to anyone its to Spain. Italy produces over 50% than Greece too. Greece is only third ranked producer (about 15% of World production) in what is a niche commodity market. Also, almost half of Greek production is consumed locally.

I do not think California is ever going to be a threat to Spain or Italy. Consumption is growing. The world is short olive oil. California can fill part of the growing needs.

If competitors should arise, it would be more the Southern Mediterranean countries if they ever get their acts together and increase quality and quantity.

At the top end side of the business, a niche within a niche, nothing is ever going to compete in terms of prestige and quality with oils like Nunez de Prado's Flor or some Italian oils or even some obscure French ones (which are a niche within a niche within a niche).

Olive Oil, an Eyewitness Guide is a great book about olive oil. Most of the other books on olive oil are more about showing beautiful pictures of Mediterranean landscapes. This one talks about cultivars and brands.

# Daylight Savings and the Market, from Victor Niederhoffer

December 3, 2012 | 2 Comments

Here's an interesting article saying that daylight savings time is bearish because it's disruptive. I believe the study should be generalized to all disruptive events.

## Vic Niederhoffer continues:

That's the point. Daylight savings time comes too rarely for a scientific study to be based on it, and there are too many comparable events that occur once or twice a year. Or perhaps I don't get the joke?

## Bruno Ombreux writes:

I cannot think of anything more disruptive than the dreaded EFA calendar in the UK power market. You may want to study its impact on spark spreads and UK-continent spreads.

Also, their days run from 23:00 to 23:00 instead of 24:00 to 24:00 like everywhere else in the World.

# French Markets and French Politics, from Bruno Ombreux

September 2, 2012 | 1 Comment

My friend was telling me the other day that he thought it was likely France would try to nationalize its banks soon enough because "[a] previous Socialist government tried (and did, which is a big reason France became an economic basket case) and Hollande certainly suggested he might do the same if elected".

My view is that France will *not* try to nationalize its banks. The last socialist government (1997-2002) did not nationalize anything. In fact, the last socialist government privatized more companies than any other government in the history of France. Hollande never suggested he would nationalize banks. What Hollande campaigned for is a separation of investment banking / trading from traditional commercial banking / deposit activities. This would be a French version of the Glass-Steagall act. There is nothing socialist in Glass-Steagall. The conservative British government is going to do the same in the UK. What remains to be seen is if Hollande will implement this reform. Because the powerful banking lobby is fighting it tooth and nail.

As far as the impact of French politics on the markets is concerned, my opinion is that they have zero impact. France is too small a country to matter economically. And 90% of the regulations are originated in the EU, that is in Bruxelles, not Paris. So you can do like me, ignore French politics except for fun.

# Getting Into Tea, from Bruno Ombreux

May 9, 2012 | 2 Comments

I have recently started to get into tea. I bought a few yixing pots and I am doing the whole gongfu cha thing.

This is a very interesting area, because tea is even more complex than wine. I am getting used to it, but I just read something which is so funny and serious at the same time, that I would like to share it.

With wine, I am used to strange comparisons. For instance, some wines are described as having a "flintlock" flavor. And they do.

Mineral impression makes up bulk of flavor. Mixed with liquor aroma it is highly reminiscent of the taste and smell of the air on a cold foggy summer morning on a beach on Mendocino's coast. I suppose Monterey is similar, but the beaches tend to be a tad coarser sand and the combined smell of cyprus and redwood is a bit more prevalent farther north.

This goes a bit beyond wine addicts descriptions. And this is not snobbery, this is actually sincere and authentic.

## Leo Jia replies:

I concur with Bruno on teas.

I drink tea everyday and have drunk many varieties: black, red, green, white, pu'er, oolong, herbal etc. Yes, one can find a lot of senses from them. One even can find different senses of the same tea at different moments.

Through the experiences with teas, coffee, wines, etc, I realized that one actually can derive sophisticated senses and feelings from just about anything one encounters in life. Be it water (from different streams, different environment/surroundings, different altitudes, and at one's different moods, etc), and likewise, air, earth, color, sound, bread, rice, etc. Very amazing.

But more amazing I think is that the very senses one gets from the things reflects, manifests and realizes one's own life. It is the fact that we can sense about these various things that makes us alive. It is also true to me that how much senses one gets actually indicate how much life one has lived - the more one has been into life, the more one can sense about the things.

With that, I am always curious about what senses I don't get, because what I don't get are what I have not got and hence are unknown to me. I am always joyful at the very moment when I suddenly sense something new, because that means I have just got something more in life.

# Dysfunctional Power Markets, from Bruno Ombreux

April 18, 2012 | 1 Comment

The gaming of the market by power utilities is described in more academic terms in Hirschhausen und Zachmann in this paper

Cf. Figure 1, strategy of withholding.

The problem is that it is very difficult to detect and prove that a utility is gaming the market. If one wants to find technical reasons to shut down a plant, they will always find technical reasons. This is just an example and there are other ways utilities can game the market. […]

The paper states that RWE, EON, EnBW and Vattenfall account for 85% of total production, and allege that these producers are engaging in practices that represent a "problem." Oddly, the authors fail to calculate (or even cite) the classic HHI "Herfindahl Index," which is the standard methodology by which US regulators apply anti-trust law to industries and mergers.

I submit that the "money quote" in this paper is:

Thus, even though demand has risen, generators have reduced capacity by 4.2 GW (1.3 GW of new construction 7vs. 5.5 GW of plant closures). 3.7 GW of the retired power plants had low generation costs. The European Commission also suggests extensive inefficiency of the existing capacity: mid-load power plants have relatively low load factors (30-40%) while several more expensive power plants show load factors of 70-90%.

It doesn't take a lot of brain power to see that rising demand and falling supply means higher prices. That's not collusion. That's good old-fashioned supply and demand. They also suggest that generators are intentionally shuttering "low-cost" capacity for the sole purpose of raising prices. That belief defies rational logic. Something else must explain that behavior. Admittedly, I don't know anything about electricity generation in Germany. So I'll ask the following simple questions:

1) E.ON's ROE from 1992 to the present has averaged about 12% — with unremarkable profitability. So if they are extracting monopoly profits, they're not very good at the game.In contrast, RWE's ROE from 1993 to 2003, averaged about 15%. But from 2004 to 2010, it averaged about 20%. So, something seems to have structurally changed for them in 2004. What was it? And why isn't E.On playing that game?

2) If there are excess profits to be made, what keeps out new entrants from eventually entering the generation market? This is especially true since the authors acknowledge the availability of long-term supply contracts from producers.

3) Price-spikes are annoying, and power-outages are troubling (especially when you're in the elevator), but the authors don't suggest that the grid has become less reliable. They just say that the price has gone up. And rising prices (absent obscene profitability) could easily be attributed to other regulatory effects. It could also be attributed to better grid reliability…

Just some food for thought from someone who is naturally dubious of blaming the evil speculators and profiteers…

# Rollover, from Victor Niederhoffer

March 28, 2012 | 4 Comments

Having considered the rollover for many years, I conclude the best thing is not to roll over at all.

## Bruno Ombreux agrees:

You are right. But that is if you have a choice. Sometime you have hedges that need to be rolled over. And it is not a choice you make. The hedges are a consequence of your underlying business, which is where you make the money. Then the hedges and the rolls are best seen as a cost, even if sometime they turn out a profit.

## Gibbons Burke writes:

An alternative to creating a single continuous contract to model the behavior of a trading regime which may holds a position across contract deliveries (as this must be tested) is to test that model's behavior using individual contract histories as you would do in real time, rolling your position at the indicated times as necessary. If you need more history for your indicators than the new contract has, then you can create a back adjusted contract anew each time with the contract inn which you have the current position reflecting the actual prices at which that contract traded, but the historical data has values from earlier contracts. This minimizes the distorting effect of cumulative rollover adjustments that you get when you make one continuous series covering the entire testing period.

## Rocky Humbert writes:

There is a paradox in this discussion. As a *theoretical* matter: I can own a cash position in something for X months/years (as a speculation, hedge or investment.) Or I can buy a future that expires in X months/years. If the p&l between the two is materially different (after taking account of leverage and financing), then this is a pure arbitrage.However, the arbitrage is problematic to exploit in physical commodities because of the logistics involved in owning and storing physical commodities. But the arbitrage should be easy to exploit in things like Stocks, Bonds, Gold, currencies, etc. The arbitrage CAN arise in the course of business precisely because hedgers, investors and speculators all have different motivations. But the arbitrageur will benefit from this dichotomy if his analysis is correct. Let's not fool ourselves: The RATIONAL ECONOMIC ARGUMENT MUST BE: the futures price is the BEST indication of where the price will be in the future. Whether that future is one month or 100 months. Any other interpretation leads to a break down in core economic principles. The rolls are simply a discontinuous manifestation of this phenomenon.

## George Coyle writes:

I am sure this is flawed logic and welcome analysis/criticism, but all this talk has me thinking stocks are a more ideal vehicle for true trend following (vs futures). No rolling/transaction costs, potential dividend yield. You don't get the leverage and are probably subject to reg T on stocks but that may not be a bad thing.

Another reason stocks are less susceptible to trend following strategies relative to futures markets are laws forbidding insider trading. The prohibition on a profits from privileged particulars prevents their percolation into prices until promulgated publicly. The predictable result is that when new information is released, it is immediately reflected in the price, causing a quantum move to the new value level, a trend exploitable by only the extremely nimble, or knowledgeable scofflaws.

No such prohibitions prevent futures traders from trading on inside information. The market exists mostly for the benefit of insiders. When they act on information they have, with their fingers on the pulse of the fundamentals of the commodity supply situation, and the condition of crops, etc., that telegraphs that information into the price. As the information spreads, and more traders act on it, the trend to the new value level which reflects the full discounting of that new data. So, the speed with which valuable fundamental data about commodities futures markets gets integrated into price slowly enough for a trend to form in price which is more than just noise. This creates enough beyond-noise trends which makes a trend following system able to operate and squeeze a profit out.

The for trend followers problem comes when the number of trend followers swells, and they all pile onto the signal - the systems acting on smaller noisy trends create their own noise and the increased noise increases the risk to the point where the real trends based on real changes in the supply-demand situation are not big enough to overcome the cost of catching the smaller losing noisy trends for small choppy losses.

# Smith Did GS a Great Favor, from Dan Grossman

March 17, 2012 | 3 Comments

I am so removed from Wall Street that this may be an obvious point:

I think it will turn out that Greg Smith did Goldman Sachs a great favor. No amount of purposeful PR could have helped GS so much and turned the tide running against GS so effectively as Smith's pompous, self-serving and unsupported resignation op ed.

Except among the irrational haters of wealth and speculation, Smith's op ed will wind up generating sympathy for GS, and I predict this week will mark the bottom of GS both in reputation and stock price. It will be pretty much all up from here.

The true criticism of GS, of course, would be its corrupt, crony-capitalist relationships with current and prior Presidential Administrations. But that's too subtle and knowledgeable a criticism. Rather the criticism in the popular mind is "greed". Smith's attempt to cloak his resignation in anti-greed will be seen through and will lead to greater acceptance of a beleaguered GS just trying to go about its business of making Wall Street work.

## Rocky Humbert writes:

As a GS alum, I would like to offer a few observations, without directly commenting on Dan's point.

When I left GS as a vice president in 1989, GS was run by Whitehead and Weinberg, successors to the legendary Gus Levy. The firm was a private partnership, and importantly, the investment banking/capital markets side of the company dwarfed the trading side of the company. This is a critical distinction from today. Sure, Bob Rubin's risk arb desk was hugely profitable. Sure, we did some big block trades in equities; but the much higher commissions of that period, and the firm's limited capital, ensured that the focus was on flow and not on principal transactions. By then, Traders were second class citizens versus the hermes-wearing, first-class-flying I-bankers who, at that time, would never ever represent a company in a hostile takeover. Of course there were some guys who pushed the envelope on occasion (I won't name names), but there was a distinct belief that everything flowed from the profitability of the clients. For an analogy of the inherent tensions between Ibanking and trading, revisit the Gluckman/Peterson feud at the ancient Lehman Brothers (pre-Amex deal).

That really was the GS culture back then. Heck, Weinberg drove a crappy Ford sedan because we did the Ford IPO. And few things could get you in trouble faster than talking badly about an important client. It was unthinkable that we would push a client into a security that we thought would turn out badly. We looked down our noses at Bear Stearns and the other bulge bracket firms who were known for that sort of thing. (Aside: I posit that the GS cultural evolution can be gleaned from the type of car the CEO drove.)

The world evolves, and I believe that the evolution of GS into its current form is a reflection of:

1) The end of its being a private partnership — which ensured risk taking with OTHER people's money. I still remember having a particularly bad losing day when Eric Sheinberg walked up to me, whacked me on the head and said with a reassuring smile, "Don't sweat it. It's ONLY money…..and it's MY money."

2) The domination of trading profits versus investment banking revenues. Management realized you can only grow investment banking to a certain size due to its service nature; whereas you can compound capital by investment and trading in a theoretically unlimited way.

3) The growth of trading technology and impersonalization of counterparty relationships. (It's much easier to "screw" someone who you don't know.)

4) The 10 percent rule, where they fire the worst performing 10% of employees every year. Back in the Whitehead/Weinberg day, such a concept would have been unfathomable. It really was a family lifetime employment sort of feel, not dissimilar to GE before Jack Welch and IBM before Lou Gestner.

5) And many other examples that correlate with a 30 year bull market in debt as a pct of GDP.

I am not lamenting here. I am simply saying that Smith is right when he observes that the GS culture has changed.

Too, the world has changed.

And, to be honest, I don't really understand why Smith wrote that piece except as an attempt to be Michael Lewis-esque, but without the chuckle factor.

## Jack Tierney writes:

Notes of interest in the GS "time to buy?" discussion: Goldman's full-year net income hit a record \$13.4 billion in 2009, then slipped to \$8.4 billion in 2010 before tumbling to \$4.4 billion last year. Goldman's share price has plummeted from its 2009 high of \$192 to the current quote of \$111. During 2009 and 2010, Goldman spent 71% of its net income buying back its stock. But last year, the company spent 264% of net income buying its stock (excluding the repurchase of preferred stock from Warren Buffet, Goldman still spent 140% of its net income buying its own shares last year - double the rate of 2009-10.) Last week, Goldman executives cashed in \$20 million worth of stock that had been "locked up" for the last three years. Over the last five years, Goldman's management spent \$21 billion of the shareholders' capital buying GS stock in the open market at an average price of \$171 a share. Today, the stock sells for \$111. On a mark-to-market basis, therefore, Goldman's stock buy-back "investment" has produced a loss of about \$7.3 billion for shareholders…. Last week, nine Goldman insiders sold their stock as fast as the law would let them. They cashed out \$20 million worth of stock at an average price of \$107.44.

## Fred Crossman replies:

Great points, Jack, on buy backs. I noted that American retailers have continually expanded at a much greater rate than the population growth. In addition to declining per store sales and income these retailers have been furiously buying back stock since 2007 to goose earnings. LOW has reduced shares outstanding by 12%, BBY 18%, HD 20%, KSS 11%, WMT, 15% and SHLD 29%. All buybacks above book value (destroying share holder value). Especially HD, now trading at 4.1 times book.

## Bruno Ombreux writes:

There is a very simple way not to be screwed by GS, or anybody else. I am talking about trading, not corporate finance.

If you are making trades directly with GS, you are presumably a company, not some small private speculator. So you have a tool which is called "Risk management policy" and you make it a sackable offense not to comply with it. In the risk management policy, you list the markets and the instruments people are allowed to trade.

For instance:

- only markets with at least 3 active market makers and x trades/per day
- only vanilla instruments like swaps In addition, you have procedures like "trader must obtain 3 quotes from 3 different counterparties prior to making a trade", and a track record of the consulted counterparties and their quotes must be kept in the trading system, for each trade. In these types of market, you are not trading every 5 minutes, so you have the time to do all this.

There is no way you are getting screwed if you restrict yourself to simple instruments and they have the best bid/ask available among several other market makers.

Sorry, but I don't understand your distinction between trading and investing. I also don't understand your definition of vanilla. I am however a fan of "rocky road" flavor.

I agree with you that entering trades that you are not sure to be able to exit is risky. But if the market provides you with a sufficient liquidity premium, it's rational and it can be profitable. But only if you do it right of course.

## Bruno Ombreux replies:

Trading vs investing: this could be the beginning of an endless semantic debate.

But let's use a couple of examples:
- trading: I buy a basket of stocks this morning with the intention of reselling before the close
- investing: I build a portfolio of stocks with the intention to keep it a relatively long time, because I think that these stocks value will increase due to whatever reason, growth, value, the economy…

I also like the following classification, which I believe comes from Minsky:
- Profits on the position neither depend on price variation of the asset, nor on cost of carry: I am investing.
- Profits do not depend on price variation, but only on positive carry: I am trading.
- Profit depend on price variation of the asset: I am speculating.

The example and the definition are not equivalent, but they give a rough idea of what trading is and what investing is. The border between both activities can be blurry. But if you invest, you do not need a market. You can buy a bond with the intention of holding it to maturity. If you trade, you need a market to close the trades.

Now, to answer your second question, what is vanilla? Vanilla is anything that is simple, easy to understand and commonly traded. In the energy markets, everybody trades swaps and Asian options. These are vanilla. What is not vanilla would be a double-barrier option on Singapore 180 cst Fuel Oil, settled at the average CAD/EUR exchange rate lagged 3 months vs the Fuel oil averaging period. That is not vanilla, and definitely more simple than many equity derivative deals.

But if you invest, you do not need a market. You can buy a bond with the intention of holding it to maturity. If you trade, you need a market to close the trades.

I will let those wiser than myself comment on the rest of your analysis, but the above jumps out at me as a poor definition of investing. Holding a bond to maturity may be a valid example of your argument, but there are plenty of people arguably INVESTING in other instruments who need a market to close their positions. A few off the top of my head include real estate, stocks, bonds not held to maturity but still held as investments, commodities including physical ones held in safes or other venues. Of course you need a market to close out most investments! I may be missing something but this seems obvious. If you cannot find someone else to buy or sell your investment at the time of closing the position, you have zero liquidity and for all intents and purposes zero value if you need that liquidity immediately. Without a secondary market, most investments cannot realize their value.

# The End of Movies, from Bruno Ombreux

February 29, 2012 | 2 Comments

Movies are now competing with other forms of entertainment. The video game industry has been bigger than the movie industry for quite a few years now. Movie theaters, Broadway, non-VOD TV, printed press… these are all on the way to obsolescence. Interestingly, the only thing from the past that is not getting superseded by newer alternative media is the radio. One can listen to the radio while doing something else, which is a great advantage.

My theory is that if something survives among my list of old entertainment vectors, it will be the radio. The idea is that people who drive cars end up listening to the radio. So there will always be some demand as long as people drive cars. I do not drive cars, but I take the cab, and cabs always have the radio on.

This is not really my personal idea, it is the result of a brainstorming session with my 22-year old trainees. The Oscar discussion on the dailyspec prompted me to organize an impromptu brainstorming session with my trainees. When it comes to the future, I trust the opinion of young people much more than 30+ years old farts.

Radio is yet difficult to replace when you are driving. All the other media can be dis-intermediated (movie theaters, Broadway, TV, dead-tree press). There is a need for content production, but not for a specific distribution vehicle (like a movie theater), or editorial effort (eg VOD and news aggregators get rid of the need for channel-imposed programs and editorial choices). Which is good news, because we are getting rid of the middlemen and of one of the means for sheeple thought-control.

# Ten Steps One Should Take to Become a Successful Speculator, from Victor Niederhoffer

I am often asked what ten steps one should take to become a successful speculator.

I would start by reading the books of the 19th century speculators, 50 Years in Wall Street, The Reminiscences of a Stock Operator by Markman, and others.

Next I would read the papers of Alfred Cowles in the 1920s and try to compute similar statistics on runs and expectations for 5 or 10 markets.

Third I would get or write a program to pick out random dates from an array of prices, and see what regularities you find in it compared to picking out actual event or market based events.

Fourth, I would read Malkiel's book A Random Walk Down Wall Street and update his findings with the last 2 years of data.

Fifth, I would look at the work of Sam Eisenstadt of Value Line and see if you could replicate it in real life with updated results.

Sixth, I would start to keep daily prices, open, high, low, and close for 20 of so markets and individual stocks and go back a few years.

Seventh, I would go to a good business library and look at the old Investor Statistical Laboratory records of prices to see whether it gave you any insights.

Eighth, I would look for times when panic was in the air, and see if there were opportunities to bring out the canes on a systematic basis.

Ninth, I would apprentice myself to a good speculator and ask if I could be a helpful assistant without pay for a period.

Tenth, I would become adept at a field I knew and then try to apply some of the insights from that field into the market.

Eleventh, I would get a good book on Statistics like Snedecor or Anderson and be able to compute the usual measures of mean, variance, and regression in it.

Twelfth, I would read all the good financial papers on SSRN or Financial Analysts Journal to see what anomalies are still open.

Thirteenth, of course would be to read Bacon, Ben Green, and Atlas Shrugged.

I guess there are many other steps that should be taken that I have left out especially for the speculation in individual stocks. What additional steps would you recommend? Which of mine seem too narrow or specialized or wrong?

## Rocky Humbert writes:

All the activities mentioned are educational, however, notably missing is a precise definition of a "successful speculator." I think providing a clear, rigorous definition of both of these terms would be illuminating and a necessary first step — and the definition itself will reveal much truth.

I think with individual stocks: one would have to really understand the sector, the company's niche and be able to monitor inside activity for possible impropriety. Individual stocks can wipe out: Bear Stearns deflated from \$60 to \$2 in no time at all. In my opinion: there is no bullet-proof technical approach, applicable to an individual enterprise situation.

A widely-held index, currency cross or commodity is an entirely different arena. And where the instrument can freely move around the clock: there will be a lot of arbitrage opportunities arising out of the fact that a high percentage of participation is inefficient, limited in both the hours that they commit and the capital they commit between time-zone changes. Small inefficiencies can snowball into huge trends and turns; and given the leverage allowed in those markets - live or die financial opportunities are ever present. So technicals overpower fundamentals. So far so good.

Comes the tricky part: to adopt statistics to the fact of unprecedented centralized meddling and thievery around the very political tops. Some of the individual market decrees may be painfully random: after all, pols are just humans with their families, lovers, ills and foibles. No statistical precedent may duly incorporate such. Plus, I suspect most centralized economies of current decade may be guilty of dual-bookeeping. Those things may also blow up in more random fashion than many decades worth of statistics might dictate. Don't tell me that leveraged shorting and flexionic interventions existed even before the Great Depression. Today's globalization, money creation at a stroke of a keyboard key, abominable trends in income/education disparity and demographics, coupled with general new low in societal conscience and ethics - all combine to create a more volatile cocktail than historical market stats bear out. 2001 brought the first foreign act of war to the American soil in centuries. I know that chair and others were critical of any a money manager strategizing around such an event. But was it a fluke, or a clue: that a wrong trend in place for some time will invariably produce an unexpected event? Why can't an unprecedented event hit the world's financial domain? In the aftermath of DSK Sofitel set-up, some may begin imagining the coming bank headquarter bombing, banker shooting or other domestic terrorism. I for one envision a further off-beat scenario: that contrary to expectations, the current debt spiral will be stopped dead. Can you imagine next market moves without the printing press? Will you find statistical precedent of zooming from 2 trillion deficit to 14 trillion and suddenly stopping one day?

Very generous post, thanks Victor…

I would add, in this day and age, learn tough typing and keyboard skills for execution and your way around a keyboard, so you don't wipe off a months profit in the heat of battle. I would also add, learn ways of speed reading and information absorption, though these two may be more "what to do before you start out".

## Gary Rogan writes:

Anatoly, I don't think really understanding the sector and and the niche is all that useful unless one knows what's going on as well as the CEO of the company, which means that in general understanding quite a bit about the company isn't useful to anyone without access to enormous amount of information. It's the subtle, little, invisible things that often make all the difference. There are a lot of people who know a lot about pretty much any company, so to out-compete them based on knowledge is usually pretty hopeless. It is nevertheless sometimes possible to out-compete those with even better knowledge by sticking with longer horizons or by being a better processor of information, but it's rare.

That said, it has been shown repeatedly that some combination of buying stocks that are out of favor by some objective measure, possibly combined with some positive value-creation characteristics, such as return on invested capital, do result in market-beating return. Certainly, just about any equity can go to essentially zero, but that's what diversification is for.

In the commodities markets it's essential to cultivate commercials who trade the same markets as you(especially in the grains.) One can glean much information from a commercial, information like who's buying. who's selling, who's bidding up the front month, who's spreading what, who's buying one commodity market and selling another, etc. When dealing with a commercial, be sure to not waste his time and have some valuable information to offer as a quid pro. Also, one necessary skill to develop is to determine how much of a particular commodity is for sale at any given time…. That skill takes a lot of experience to adequately gauge the market. Also, in addition to finding a good mentor, listen to your elders, the guys who have been successful speculators for decades, the guys who have seen and experienced it all. Avoid the clerks, brokers, backroom guys, analysts, touts, hoodoos etc. Learn to be cold blooded and be willing to take a hit, even if you think the market might turn around in the future. Learn to avoid hope, as hope will ultimately kill your bankroll. When engaged in speculation, find one on one games like sports, cards, chess, etc that pit you against another person. Play these games aggressively, and learn to find an edge. That edge might translate to the markets. Still, while being aggressive in the games, play a thinking man's game, play smart, and learn to play a strong defensive game……a respect for the defense will carry over to the way you approach the markets and defend your bankroll. Stay in good physical shape, get lots of exercise, eat well, avoid excesses.

Given that manipulation is still prevalent in some Asian markets, I would add that, for individual stocks in particular, one needs to  understand manipulators' tactics well and learn to survive and thrive under their toes.

## Bruno Ombreux writes:

Just to support what Jeff said, you really have to define which market you are talking about. Because they are all different. On one hand you have stuff like S&P futures with robots trading by the nanosecond, in which algorithms and IT would be the main skill nowadays, I guess. On the other hand, you have more sedate markets with only a few big players. This article from zerohedge was really excellent. It describes the credit market, but some commodity markets are exactly the same. There the skill is more akin to high stake poker, figuring out each of your limited number of counterparts position, intentions and psychology.

I note that the Chair ignored my request to precisely define the term "successful speculator," perhaps because avoiding such rigorousness allows him to define success and speculation in a manner as to avoid acknowledging his own biases. I'd further suggest that his list of educational materials, although interesting and undoubtedly useful for all students of markets, seems biased towards an attempt to make people to be "like him."

If gold is up a gazillion percent over the past decade, and you're up 20%, are you a successful speculator?If the stock market is down 20% over a six month period, and you're down 2%, are you a successful speculator?If you have beaten the S&P by 20 basis points/year, ever year, for the past decade, without any meaningful drawdowns, are you a successful speculator?If you trade once every year or two, and every trade that you do makes some money, are you a successful speculator?

If you never trade, can you be a successful speculator?

If you dollar cost average, and are disciplined, are you a successful speculator?

If you compound at 50% per year for 10 years, and then lose everything in an afternoon, are you a successful speculator?

If you lose everything in an afternoon, and then learn from your mistake, and then compound at 50% for the next 10 years, are you a successful speculator?

If you compound at 6% per year for 10 years, and never have a meaningful drawdown, are you a successful speculator?

If the risk free rate is 6%, and you are making 12%, are you a more successful speculator then if the risk-free rate is 0% and you are making 6%?

If you think you are a successful speculator, can you really be a successful speculator?

If you think you are not a successful speculator, can you be a successful speculator?

Who are the most successful speculators of the past 100 years? Who are the least successful speculators of the past 100 years?

In conjunction with the chair's mention of valuable books and histories, I would append Fred Schwed's Where are the Customers' Yachts?.

While ostensibly written with a tongue-in-cheek hapless outsider view of 1920s and 1930s Wall Street, it has provided as many lessons and illustrations as anything by Henry Clews. In this case, I am reminded of the chapter in which Schwed wonders if such a thing as superior investment advice actually exists.

## Pete Earle writes:

It is my opinion that the first thing that the would-be speculator should do, even before undertaking the courses of actions described by our Chair, is to open a small brokerage account and begin plunking around in small size, getting a feel for the market, the vagaries of execution quality, time delays, and the like. That may serve to either increase the appetite for such knowledge, or nip in the bud what could otherwise be a long and frustrating journey.

The obligatory Wikipedia* definition of speculation is investment with higher risk:

Speculating is the assumption of risk in anticipation of gain but recognizing a higher than average possibility of loss. The term speculation implies that a business or investment risk can be analyzed and measured, and its distinction from the term Investment is one of degree of risk. It differs from gambling, which is based on random outcomes.

There is nothing in the act of speculating or investing that suggests holding times have anything to do with the difference in the degree of risk separating speculation from investing

By this definition one must define risk and decide what comprises high and low risk — which may be simple in extreme cases but (as we have seen repeatedly) is not very straightforward in financial markets

*Chair is quoted in the link

## Alston Mabry writes in:

I'm successful when I achieve the goals I set for myself. And rather than a target in dollars or basis points or relative to any index or ex-post wish list, those goals may simply be to act with discipline in implementing a plan and then accepting the results, modifying the plan, etc.

And don't forget Ed Seykota: "Everyone gets out of the market what they want". I find that everyone gets out of life what they want.

Plenty a market participant is not in it to make money. Fantastic news for those who are!

## Bruno Ombreux writes:

This will actually bring me back to the question of what is a successful speculator.

In my opinion success in life is defined in having enough to eat, a roof, friendships and a happy family (as an aside, after near-death experiences, people tend to report family first). You can forget stuff like being famous, leaving a legacy or being remembered in history books. If you are interested in these things, you have chosen the wrong business. Nobody remembers traders or businessmen after their death except close family and friends. People who make history are military and political leaders, great artists, writers…

So you are limited to food, roof, friends and family. Therefore my definition of a successful speculator is a speculator that has enough of these, so that he doesn't feel he needs to speculate. I repeat, "a successful speculator does not need to speculate."

I simply think that a successful speculator is one who makes money trading. Among soccer players Messi, Ibrahimovic are considered very successful. They consistently score. They experience short periods without scoring. Similarly, traders should have an equity line which consistently prints new highs with low volatility and a short time between new highs. Like soccer players and other athletes it is their mental characteristics the main edge rather than knowledge of statistics. One can learn how to speculate but without talent cannot play the champions league of traders and will print an equity line with high drawdowns struggling losing too much when wrong and winning too little when right. Before dedicating time to find a statistical edge in markets one should assess his own talent and train psychologically. In this regard I like Dr Steenbarger work. In sports as in trading you very soon know yourself: your strengths and weakness. There is no mercy. You are exposed and naked. This is the greatness and cruelty of markets and competition. This is the area where one should really focus in my opinion.

## Steve Ellison writes:

To elaborate a bit on Commander Pezzutti's definition, I would consider a successful speculator one who has outperformed a relevant benchmark for annual returns over a period of five years or more. Ideally, the outperformance should be statistically significant, but market returns can be so noisy that it might take much of a career to attain statistical significance.

## Jeff Rollert writes:

I propose a successful speculator dies wealthy, with many friends. Wealth is not measured just in liquid terms.

Should a statistical method be preferred, I suggest he is the last speculator, with capital, from all the speculators of his college class.

In both cases, I suggest the Chair and Senator are deemed successful, each in their own way.

If I may wager my 2 cents here.

I would define a successful speculator as someone who has achieved a record that is substantially above the average record of all speculators in percentage terms during an extended period of time. The success here means more of a caliber that one has acquired which is manifested by the long-term record. Similarly regarded are the martial artists. One is considered successful when he has demonstrated the ability to beat substantially more than half of the people who practice martial arts, regardless of their styles, during an extended period of time. It doesn't mean that he should have encountered no failures during that time - everyone has failures. So, even if that successful one was beaten to death at one fight, he is still regarded as a successful martial artist because his past achievements are well revered.

With this view, I will try to answer Rocky's questions to illustrate.

## Julian Rowberry writes:

An important step is to get some money. Preferably someone else's. [LOL ]

# Some Reflections on a Trip to London and Paris, from Victor Niederhoffer

January 3, 2012 | 3 Comments

1. Every store in London was having a sale of 50% or more except for the Bates one I went to to buy a hat, and all the big stores like Harrods had queues of at least 2 hours. In Paris, no stores had sales and business seemed quite slow except for the health food stores that substitutes quinoa for rice and hummus. Why is there so much better retailing in London than Paris? Does it have to do with the service rate or National Character? The marginal utility for consumers to buy goods in London and Europe rather than property seems to be a function of the much larger ratio of space to population in US versus Europe. When you have 100 square feet to a person, goods seem very attractive and the Holidays with all their bargains, bring out in London "50% of the population". By the end of a week, people are willing to spend a lot more to buy things than at the beginning when they're still testing the waters and looking for bargains. Can this be quantified in markets?

2. The drop and close below 1200 on Dec 19, 2011 is right out of the playbook of the Trojan War. Time and time again a day before the death of one hero or another, in this case Hector as he firebrands the Greek ships and kills and wounds one Greek defender after another, including Ajax, Menelaus, and Odysseus, the Gods look down, especially Zeus, and say, "look he's going to die tomorrow, let him have a blaze of victory today before he goes to Hades as he's put up a good fight and is the favorite of a few mistresses and daughters." One receives a pretrial settlement letter from Dan about a HP executive's harassment of a party planner, including his showing her his million dollar balance at the ATM. And it gets him in trouble because there is an obvious attempt to cover up through his assistant who might not be buyable off now that he is no longer top guy. It's right out of the Trojan war where all the problems arise from romance and the fate of the war hinges on who can seduce Zeus the last, and which Goddess is consumed by revenge the most because Paris chose Aphrodite and how they can use their wiles to turn the tides of war.

3. A trip to the British Museum starts with a building cramming exhibit of Russian Architecture right after the Revolution to show the Russian's spirit and intelligence, and the love and egalitarianism of Russia by the British right now. But at the British Museum a room is devoted to Roman everyday life then compared to England today, and the conclusion is that it's pretty much the same with the soldiers being able to retire to a nice plot of land after 20 years of service then and now. But on looking closer one sees that most of the wealthy in Greek times were the freed slaves who were able to fill the everyday jobs of merchants, doctors, and financiers since they were not tarnished by striving for money and didn't possess extensive land holdings.

4. Throughout Europe the opportunity cost of time is close to zero. Queues are everywhere because free admission is given to all the attractions. One can only get into the Louvre through a back entrance as the lines at the front are 3 hours long, but when you do get in, you have to walk through 10 miles of religious paintings depicting sacrificial and revengeful scenes from the Bible. No such luck at a the Musee D'Orsay where one would have to wait 5 hours to get in, even after purchasing a ticket at the only billetiere open in Paris.

5. London is the theater capital of the world, and it's nice to see the common man at all the events, enjoying his ice cream between acts at 1/4 the price of US events. I have to walk out of Crazy for You and Matilda, two of the hits there, because the music is terrible, and the plots totally contrived and hateful to the business person. The Crazy For You plot is exactly the plot of the current Muppets movie with their depiction of the heartless business man who wishes to close down the theater and the decent poor folk who must stage a show to earn enough to buy out the theater from the evil profit mongers. I enjoyed Three Days in May which shows men as they should be with compromises between Churchill, Chamberlain, and the Hunting Saint that led to the British refusing to surrender as it becomes clear that France was going to capitulate in a week. "Neville, can I chat with you for a half hour before the meeting tomorrow. Without your support, I'll have to resign because I don't believe we should give up," Churchill said. How many times one has been in that position as every man was for himself as in this case the estimate that came back from the front was that 2000 men would be returned from the Dardanelles rather than the 250,000 that came back. Worst of all as Churchill pointed out to his cabinet dissenters, is a show of uncertainty and disharmony as the public would leap on the weakness and the whole battle would be lost. One finds the courage and diplomacy of Churchill inspiring in this case, and one did have it in his rackets career.

6. A highlight of the trip is a visit to Ile de la Cite to see the prisons where the upper class and producers were kept before being guillotined. But instead one lands at the Sainte Chappelle where one is seated in the first row of this 14th century church, to hear a medley of Renaissance music with harpsichord, viola, various flutes and a singer. The highlight as always is a Couperin and Bach piece which is invariably ingenious and beautiful compared to the predecessors. One was mistakenly given a VIP seat here as the reservation made from a fancy hotel and I am reminded of the most valuable thing I got from Soros other than the two tennis can thing. Once I had pneumonia and the hospital mistakenly heard that I was a partner of Soros and they gave me the best room in the hospital, about 2500 square feet with a beautiful view of the park. I did meet a great Dr. there, Dr. Lou Depalo, who I would recommend to anyone with a respiratory problem of any kind, who bought me a Barrons, and I bonded with when it turned out that he had a total love of the Master and Commander canon and unlike me was a nautical personage.

I was in Paris with my wife and daughters over the week preceding and including Christmas. We didn't do much shopping since it was mostly about taking the kids to the main museums, and they all know how much I hate "shopping" but we did spend a couple of ours at Galerie Laffayette, their main shopping mall, on Christmas Eve and the level of energy seemed pretty good to me, but I don't have too many comparison points. I also didn't see any sales signs, but could that be a sign of strength?

The outdoor shopping area at the lower end of Champs Elysees was so crowded in the evening it was almost impossible to walk, and this is definitely not the height of the tourist season. The faces of people on the metro which we used a lot seemed somewhat grim, but that's also hard to interpret without recent comparison points.

1. Back when I lived in the UK in the 1980's, there were semi-annual sales (post-Christmas and July). This was a tradition at the likes of Selfridges, Harvey Nichols, and the other serious London department stores. Prices were generally not discounted except during these sales. No self-respecting Londoner would shop at Harrods (except at the food court) — as it was mobbed with foreign tourists, and the prices were exorbitant. Perhaps a current Londoner can share whether the semi-annual sale pattern still exists.

2. In comparing London and Paris, one recalls Adam Smith's (and Napoleon's) observation that England is a Nation of Shopkeepers.

3. The Chair asks, "Why is there so much better retailing in London than Paris?" As Perry Mason would say, "Question assumes facts not in evidence."

1. About the traditional semi-annual sales (post-Christmas and July), it is the same in France, but the Winter sales are starting only next Wednesday. Which explains why there were sales in London and not in Paris. Different calendars.

2. About the English nation of Shopkeepers, it can be explained by different cultures too. Sales are widely attended in both countries, but from my anecdotal experience living both in London and in Paris, they are really a sacred institution in London compared to Paris.

Kathryn Schulz, in Being Wrong, one of the books on the Chair's recommended reading list, wrote:

In short, we are wrong about love routinely. There’s even a case to be made that love is error, or at least is likely to lead us there. Sherlock Holmes, that literary embodiment of our … ideal thinker, 'never spoke of the softer passions, save with a gibe and a sneer.' Love, for him, was 'grit in a sensitive instrument' that would inevitably lead into error.

# Thoughts on the Old World, from Bruno Ombreux

December 4, 2011 | 1 Comment

I have noticed that Europe has an additional dimension compared to the States.

When you travel the USA, all you see is a 3-dimensionnal landscape. In the Old World, there is a fourth dimension that is the history behind the landscape.

In Europe, when you look at a street, a village or a hill, you do not see only a street, a village or a hill. You see 2000 years of history for this street, this village or this hill. There is always a little ruin nearby.

How can I say this? There is extra depth in the Old World. It makes everything more full and life richer.

Same thing when you are marrying true blue blood. You are not marrying merely one guy or one girl. You are marrying history, a line going back centuries. You are becoming part of that line. I am talking true blue blood. Not recent ones like the fake nobles made by Napoleon and the likes.

Maybe the best way to explain it to American people, is to explain it to the ones who are independently wealthy. So if you are wealthy, have you ever tried explaining to poorer people what it felt like to have enough money not to worry about the future? Did they understand you? Or did they just looked at you as if you were a freak? Same problem with explaining the Old World additional dimension.

## Stefan Jovanovich responds:

Dear Bruno,

1. The "Old World" of Europe is not nearly as ancient as the travel brochures like to pretend. The governments of all but the most recently admitted states in the American Federal Union have longer established histories (and older unchanged borders) than any of the nation-states in Europe.

2. The "ruins" in America are there and some of them are almost as old as the catacombs; but they are not on display because what Americans have always sold Europeans is the idea of the United States as this wild, unsettled country. You can find railway posters of the Union Pacific advertising the untamed country of Yosemite to potential German and English tourists when the Ahwahnee was offering 10-course meals. Europeans have always come to the U.S. to see the "new"; that is why they still like California - it always photographs like something just unwrapped for Christmas (the best time to take the picture because the smog is being blown away from the coasts) even though it now has an industrial history as old as the English Midlands was in the 1950s.

3. There is a great deal of blue blood here, but it has one fatal defect - there are no titles to identify the "line going back centuries". There had better not be; it is against our Constitution and, if you are going to claim ancestral superiority based on Plymouth Rock and Valley Forge, you can't at the same time be spending all your time bragging about being descended from European nobility. Those claims of ancient European lineage are the very ones being made by the people whose genealogy is - shall we say - questionable. That was just as true in the 19th century as it is now. The Astors and others who were eager to acquire British class did not have family histories that could trace back to the American Revolution, let alone the founding of the Massachusetts Bay Colony and New Amsterdam. Since there were enough people around like the Roosevelts, they had no choice but to go looking for an Anglo or Franco merger.

4. I can't speak for "poorer people" in the rest of the world, but I can assure Bruno that Americans have no difficulty imagining what it felt like to have enough money not to worry about the future. The turning point in John Kennedy's campaign for the Presidential nomination came in West Virginia. A coal miner asked Kennedy if he had ever had to work for a living. Instead of offering the standard nonsense (Daniel Patrick Moynihan's "I grew up in a poor family", John Edwards' "my Dad was a mill worker" (his father ran the factory), Warren Buffett's "I had a paper route", etc.), Kennedy had the balls to say "No, I never have." The miner's reply was "Good for you." That brought down the house and ended Hubert Humphrey's ridiculous attempt to portray himself as a man of the people.

Most of foreigners' difficulty in understanding America comes from another long-established fact about the United States: the recently-arrived (usually the scholarship children of the immigrants) do almost all the public talking about the country. The oldest tradition in America is to have the A-students lecture the rest of us and tell the world at large about how we are not living up to the traditions of the Republic. (Benjamin Franklin was doing it - and worrying about the Pennsylvania Dutch, er Deutsch when the Penn family was keeping quiet and making certain their land title was secure and paying Franklin to fix it.)

It takes at least one or two more generations for the newly-arrived Americans to discover what Richard Jennings, California blue-blood member of E Clampus Vitas and author of the revised California Corporations Code, once said to our law school class at Berkeley: "Remember someone in this University is going to drop out of school or leave with a "C" average and end up making more money than the rest of us combined." What he did not add was that, while that person might be the child of recent immigrants (see Steve Jobs), the odds were much greater that he would come from a family like that of Mr. Buffett's bridge partner - one old enough that the possibility that great great grandmother may have been one of the "seamstresses" who set up shop above the water table in Seattle can be safely ignored. What I would have added is that it is far more than an even money bet that the same family will be "progressive" enough to be in favor of raising the estate tax. Preventing the newly-arrived from doing what grandfather did to escape the ravages of the tax code is perhaps the most well-established of all the traditions of the better classes of 4-dimensional Americans.

Have a wonderful holiday.

Stefan

# European Version of QE, from Paolo Pezzutti

After euro leaders announced a new (big) aid package for Greece and measures to prevent bond yields rising further in Spain and Italy, it seems that Europe has solved its sovereign debt problems…. Markets celebrate the European version of QE. Also Europeans (we'll see what Americans will now do about it, but I think the answer is pretty clear and markets know it) can now delay any tough decisions on deficits. Someone else will pay the bill. Pretty sad. However, markets go up and everybody is happy so far.

## Kim Zussman writes:

Do not recall the oft-heard warning that a Greek default or failure to raise US debt limit will result in financial Armageddon prior to the Lehman collapse. That not yet distant memory still has usable power. Perhaps a day's meal in that.

## Bruno Ombreux writes:

The 100s of billions will mainly come from the pockets of the German, French and Dutch taxpayers, since the ECB printing powers are limited. As for the Greek, they have lost their sovereignty but they will find freedom though work. You know. Arbeit macht frei. That is at least for the next few months until bailout 3 is needed and the whole show starts again.

Also, about the question from the bleachers: How are the ECB's "printing powers" more limited than the U.S. Federal Reserve's?

They need to maintain some capital and this capital is provided by the central banks of the member states. They cannot do too outrageously stupid things, because they can get bankrupt if some member state central banks stop capitalizing them.
And you can be sure that the German or Dutch have an uncle point.

Politically, the ECB is also far less inclined to print than the Fed:

- It has only one mandate, low inflation, whereas the Fed has a dual mandate: inflation and economic growth. In effect, the ECB doesn't really care about economic growth. - The German have been paranoid about inflation since Weimar and would not let the ECB go too far. - If you look at the ECB board, it is predominantly hawkish according to analysts and observers that spie every word uttered by these guys.

But the main reason they have less power to print than the Fed is that they have to please a lot of member states. Which creates checks and balances. Whereas the Fed only has to please one guy, the US president, who nominates board members.

Leaving any debates on what is considered QE and what is not.

The ECB has lent to the periphery through its rediscount window Euro 330 billion, this is in addition to the Euro 75 billion in secondary market purchases of peripheral country bonds. The ECB has a capital cushion of about Euro 10 billion. One could argue this might be a stretch of the single mandate of the ECB's 1998 charter.

To Ireland alone the ECB has exposure of Euro 180 billion, or about 100% of Irish GDP.

Other thoughts on the most recent Euro summit:

The general reaction in markets and the street research has been this plan delivers slightly more than expected and is a bold plan. Not surprisingly this has been the analysis of most of the rescue packages globally since 2008. Yet, the failure of the packages on so many levels is fairly evident if one looks at economic growth, interest rate spreads, etc.

The most recent package clearly will buy some time. How much is an open ended question but I expect much less than previous packages as the Pavlovian reaction wears thin.

Amongst the many issues of implementation, political approval, private sector etc. I think the key failings are:

1) There was no increase in the size of EFSF. Furthermore, even the relatively paltry and debatable rating of current EFSF has yet to approved. To cover Spain and Italy would require 1.5-2.0 trillion Euros.

2) Concessions on rates and maturity extensions for Ireland and Portugal are nice but small relative to the fiscal adjustment required.

3) The debt relief to Greece is insignificant and will bring debt to GDP from 172% to around 150-160%, depending on whose estimates you are using.

4) Given that there is not much debt relief and the fiscal adjustment is massive there cannot be much hope that the domestic political willingness in Greece will be there to stay the course.

# Quantitative Studies, from Newton Linchen

After 5 years or so, I finally got to the point of confidence in conducting basic quantitative studies. (Very basic…)

While reading again Philip's book "Optimal Portfolio Modeling", I got stuck in the following sentences:

"Professor Niederhoffer was just such a divergent thinker.

His help and guidance taught me to see things at their simplest. That is the essence of his approach. His enlightenment also helped me to learn how to avoid the numerous pitfalls that can arise in quantitative studies. *In fact, one of the things he taught me was what not to do on a quantitative study*."

I couldn't help to think what such advice would be…

And what the Specs thinks of what one should avoid while performing any counting studies.

## Steve Ellison writes:

Be very careful to consider only information that was known at the time. For example, when doing a study that uses the high price of the day, you cannot know that any price will be the high of the day until after the close. Similarly, you cannot act on the closing price or anything based on the closing price, such as a moving average, until the next day.

Beware of data mining bias. If you test the same set of data enough times, you will find some results that appear to have statistical significance, but occurred just by chance. For example, I analyzed the most favorable trading days of the year. There are an average of 252 trading days per year, so one would expect 12 days to have results with p<0.05 just by chance. You need to control for data mining bias either by setting a more stringent p threshold or testing out of sample. Any time you have considered multiple strategies and selected the one with the best results, you should assume that part of the good result was by luck and expect worse results going forward.

Statistical significance is not necessarily predictive. In an era of much quantitative analysis, a regularity may not last long. It has happened more often than I would expect by chance that I found a pattern that was bullish or bearish with statistical significance, and the out of sample results were statistically significant in the opposite direction.

## Bruno Ombreux writes:

Data mining bias can be experienced in the most vivid manner with the new Google correlation engine. It can come up with some of the weirdest, actually impossible, correlations. Google correlation results are more illustrative and striking than any theoretical academic stuff about multiple comparisons.

## Phil McDonnell writes:

An incomplete list of things NOT to do on a quantitative study:

1. Avoid retrospective data. Many fundamental data bases have retrospectively adjusted data. sometimes the data is adjusted years after the fact and could not possibly be known at the time.

2. Avoid retrospective price data. Many so called quants pat themselves in the back for 'correcting' their data after the fact. Any valid study must include the data as it was known at the time.

3. Avoid the part whole fallacy. There is more on this in the Chair and collab's book Practical Speculation.

4. Use non-parametric/robust statistics to avoid fat tail issues.

5. Simplify your studies to a very small number of variables.

6. Avoid looking at simultaneous relationships. They are descriptive and not tradeable. Instead concentrate on predictive relationships.

7. Avoid indexes, rather use prices that actually trade.

This list is only some of the pitfalls and traps to avoid in doing a proper quantitative study.

## Newton Linchen writes:

It has happened more often than I would expect by chance that I found a pattern that was bullish or bearish with statistical significance, and the out of sample results were statistically significant in the opposite direction.

Isn't that annoying?

Doesn't it pushes us to the other side of the coin, of pure "tape reading", etc?

# Not a Good Week for Market Efficiency, from Steve Ellison

The chart of the S&P 500 this week looked very much like what an uninformed member of the public would think a random walk looked like, but was actually a highly nonrandom pattern of repeated reversals, at least until late Friday afternoon.

The biggest move all week occurred when an easily predictable aftershock, that should by the theory have already been priced in, occurred in Japan.

## Bruno Ombreux writes:

On monthly returns for sure.

I remember when I was a MBA student in finance class the teacher wanted to show us that the market was random. He displayed a chart of the real DJIA side by side with a chart of a random walk, then asked the class to pick the real stock market. 150 people choose the random walk and 2 of us, that includes me, picked the real stock market.

By the way, you need balls to go 2 against 150 in an MBA classroom.

I think there are other things at play here. 150 people getting wrong get be attributed to the guys being lemmings, and 2 right guys can be attributed to us knowing the chart in advance (we were among a few with previous trading experience).

# Champagne as Market Indicator, from Bruno Ombreux

January 6, 2011 | 1 Comment

This is an occasion to repost my 2006 wine study:

I looked at the following time series:

- Champagne and Sparkling wines, a French CPI component from INSEE.
- 11° red wine, another French CPI component from INSEE.
- CAC40 index

In addition, I defined a "Feelgood index", which is the difference between champagne and red wine log-returns. The rationale is that when people are feeling good, they drink champagne rather than your average Joe's supermarket 11° red wine.

The CAC40 choice could be discussed. France is accounting for only 25% of champagne sales, the rest being exported. However, due to the correlations between world stock indices, this choice is not critical.

Data is monthly from 1998 to 2005 included.

The conclusion is that there is no correlation between champagne, red wine, feelgood and the CAC40 returns, either contemporaneous or lagged one month. Wine is a proper vehicle for diversification.

A second study, with different prices, those from the Association of Champagne Dealers, looking at yearly returns from 1990, also concludes that there is no correlation. Since there are only 16 yearly data points, the 95% confidence interval is huge:
-0.54 < correlation < +0.50 Then this second study can be ignored. However, it doesn't contradict the first one.

Champagne returned 22% over the period and 11° red wine 31%. These are poor returns, far below the stock market, even before storage costs. However, annualised volatilty of returns is very low: 1.3% for champagne et 1.9% for red wine.

Nota Bene:

1) This is not a study about returns from investing in top wines like Margaux or Haut Brion. This is a study of run-of-the-mill champagnes and red wines, who shouldn't commend any age, scarcity or snob-appeal premium.

2) It is neither a study about generating alpha from wine-picking. We are only looking at broad wine indices from INSEE.

——————-

Results for contemporaneous series

(lagged series give similar results)

*CAC40  vs Champagne*
Residuals:
Min         1Q     Median         3Q        Max
-0.0107351 -0.0021576 -0.0002676  0.0019569  0.0105993

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.0020951  0.0003879   5.402 5.05e-07 ***
cac         0.0010887  0.0064082   0.170    0.865

Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.003771 on 93 degrees of freedom
Multiple R-Squared: 0.0003103,  Adjusted R-squared: -0.01044
F-statistic: 0.02886 on 1 and 93 DF,  p-value: 0.8655

*CAC40 vs red*
Min        1Q    Median        3Q       Max
-0.025410 -0.001498 -0.001408 -0.001034  0.014695

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.0014074  0.0005554   2.534   0.0129 *
cac         0.0022486  0.0091757   0.245   0.8069

Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.0054 on 93 degrees of freedom
Multiple R-Squared: 0.0006454,  Adjusted R-squared: -0.0101
F-statistic: 0.06006 on 1 and 93 DF,  p-value: 0.807

*CAC40 vs Feelgood Index*
Residuals:
Min         1Q     Median         3Q        Max
-0.0141041 -0.0032415  0.0001090  0.0033706  0.0277683

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept)  0.0006878  0.0006367    1.08    0.283
cac         -0.0011599  0.0105198   -0.11    0.912

Residual standard error: 0.006191 on 93 degrees of freedom
Multiple R-Squared: 0.0001307,  Adjusted R-squared: -0.01062
F-statistic: 0.01216 on 1 and 93 DF,  p-value: 0.9124

# P/E forecasting, from Steve Ellison

January 3, 2011 | 2 Comments

You can get as-reported earnings for the S&P 500 from 1988 on at Standard & Poor's website.

Using 12-month trailing earnings for each year's September quarter (the last that would be known by Dec. 31) to calculate an E/P ratio for the S&P 500 as of Dec. 31, I get a somewhat positive correlation of trailing E/P to year-ahead returns with t=1.10, R sq=0.057, p=0.29, and N=22.

## Larry Williams writes:

My model for the DOW suggests a 12.25% growth for the year, slightly above the long term average growth.

For the S&P, I get 10.6 % barely above the long term average.

## Bruno Ombreux writes:

There are two things I don't like in P/E or E/P studies.

1) Your regression is in the form:

-1 + P(t)/P(t-1) = f[P(t-n)/E(t-n)] + e we have the same variable on both sides, and even if it is lagged I am not sure standard regression is OK to handle this type of formula. Just to give you an idea, multiplying both sides by P(t-1), it is actually P(t) = P(t-1) + P(t-1)* f[P(t-n)/E(t-n)] + P(t-1)*e

This is certainly amenable to study, but not with the standard regression toolbox.

2) Price is more volatile than earnings. There is a subtle bias introduced by the fact that over the estimation sample, high P/E will be naturally followed by lower P/E, and vice-versa. This is a bit like regression to the mean but more subtle. This can lead to spurious mean-reversion.

The issue is not really the dependent variable. It is using the Shiller variable with its serial correlation. One way to use the Shiller variable would be to take every tenth month. That might work but you would have one tenth the data. You still might have the Holbrook Working flaw because of the averaging. The averaging also leads to the Slutsky-Yule effect which creates spurious sinusoidal artifacts in the adjusted variable when no such sinusoidal effect is actually present in the original data..

# How Not To Cheat On Your Wife, from Bruno Ombreux

August 10, 2010 | 5 Comments

Actually, now that I think about it, this dating stuff is very much like trading. You can be a scalper or a buy-and-holder.

I would advise young guys to be scalpers. Buying-and-holding can produce long term satisfaction. But it can also produce terminal drawdowns. Odds are 50/50 that it is one way or another.

Scalping, on the other hand, never has big upside or downside. But the little upsides can add up quickly. There is no downside actually, because contrary to trading, you cannot lose. If she says no, you gain or lose zero. If she says yes, you gain a small something. And you only need a small success rate to be successful. Go out once a week, then ask 5 girls. If only 1 in 20 says "yes", it is still more than 1 girl a month. Go out twice a week, and it is more than 2 per months.

In practice, we are not looking at constant rates. There is some variance. Spend 15 days at Ibiza in summertime. That's 15 girls. Then you can get into a 1-year relationship, that's 1 girl in 1 year.

Why do I advise young guys to be scalpers? Because it makes for an happier life later. Since you will have done it all in your misspent youth, you won't feel the need to cheat on your wife. And when you reach old age, you will not go through the all-too-common crisis, buy a Harley, chase young skirts. You won't need to, because you will be thinking that you have not wasted your youth and you won't need to prove anything to yourself.

There is a French saying: "Les jeunesses sages font les viellesses dissolues". Which can be roughly translated: well behaved young men make unbehaved old men.

# Data Mining at Stock Trader’s Almanac, from Sam Marx

In the 2010 Stock Trader's Almanac on pg. 52, there is an article entitled "Take Advantage of Down Friday/Down Monday Warning".

The gist of the article is that because of the importance of Friday and Monday as trading days a down Friday followed by a down Monday is Bearish.

It seems logical and they printed the data to "substantiate" this conclusion. Their table of data indicates the market is down anywhere from 18 to 54 days later.

I wanted some more immediate and more definite results that I might use, so I tested where is the market 5 days after a Down Friday/Down Monday for the years Feb. 1993 to July 2010 .To my surprise I found that on average the market is not down but up 5 days later and this occurred approximately 60% of the time.
I tested it further for 10, 15, 20, 25, 30, 35 days after a Down Friday/Down Monday.

With the exception of 25 days later, the market was up all those other days.

The market was slightly down on the 25th day but was up again, slightly 30 & 35 days later.

I have not broken up the test into smaller time segments to check the hypothesis further because I was truly disappointed.

Just the numbers 18 and 54 are suspect. They smell of cherry-picking.

But there is an even bigger flaw. It is that it doesn't make any sense. One could forgive people not knowing about multiple hypothesis issues. They require education. But throwing away common sense as they do, is a different matter.

Why would something that happened today impact the market 18 to 54 days later? And not in between? When we all know that if there are dependencies they are very weak and vanishing with the passage of time?

# Behavioral Game Design, from Bruno Ombreux

Here is a great article on Behavioral Game Design from a Microsoft game designer. Since trading has essentially become a computer game…

Fascinating analysis! Note that you can insert "customer" or "employee" wherever the article uses "player", and the insights are just as true and powerful.

# A Pleasant New Food, from Victor Niederhoffer

May 4, 2010 | 2 Comments

It's not barbecue, but today I had the pleasure for the first time of tasting lilacs. I was inspired by a kid's book that said that to find out why bees like the taste of sweet things you should squeeze the nectar of some nettle plants. I took Aubrey to the Bronx Botanical Gardens and we smelled the lilacs. By far the best-smelling were the hyacinth lilacs. To me it's the best smell in the world, and it stops time and elicits every romantic spring personage that one could ever imagine.

Inspired by the reverie, I couldn't resist tasting a number of the small flowers. I found that the white ones on the top of the trees were superior in taste and smell to the red ones, especially lilac poincare and lilac common. The taste is like a mixture of raspberries and sweet peas. A slight tartness at first, but then a beautiful saladly green with sweet overtones.

I've seen some recipes for lilacs subsequently, but surprisingly nothing on the actual taste of lilacs, indeed almost a googlewhack. I highly recommend that all speculators take a break from their trading before or after the daily fray and sample a few in their area.

## Phil McDonnell writes:

As a home gardener I have been amazed at the number of flowers that can be eaten and are considered delicious gourmet treats. For example, zucchini flowers. Zucchini plants have two types of flowers, male and female. The male flowers stand erect and tall as is only proper. The female flowers are short stemmed and demurely lower. Even though the two flowers look similar, I find it very easy to remember the difference.

The taller male flowers are the first to be found by the bees. They next visit the lower females, thus facilitating fruiting. If bees are lacking, then the higher male flowers can pollinate the lower female flowers simply through wind action. After the male's job is done the gourmet can harvest the male flowers for a real treat. Just sautee in butter, salt, garlic and onions. Each female will give you a zucchini.

Reading Vic's post, I was thinking exactly on that: "fleur-de-courgette" and I didn't know it translated as "zucchini flowers". Thanks Phil. I'll be growing some in my potager every year. Beignet-de-fleur-de-courgette is a terribly good (and not so difficult) flower recipe. Here is a video.  I expect to have my own olive oil in the coming years to made it 100% with raw home products.

## Bruno Ombreux writes:

May I recommend a French delicacy: Crystallized violet flowers.

They are very easy to make. These are the flowers you use.

# Dangers of Sniping, from Nick White

When discussing today's market action with a friend, the following chat conversation took place:

me: Market's crazy again.

friend: Yep i noticed…herd mentality causing kids to run from one side of the playground to the other.

me: Fine by me. Opportunity to play sniper in the market again and take advantage of people being dumb.

friend: one of the hardest things about being a sniper is not neccearily the distance from which you shoot, or the accuracy required when you shoot, or finding that perfect location, it's probably being too focused through the sights of your scope to not notice the private that snuck away went arround you and is now standing behind you with a gun to your head.

Wise words indeed…

## Ken Drees writes:

That's why snipers work in teams–to watch prey and to cover rear guard.

I was told that the spotter is not technically needed. He is only there to share the psychological burden arising from killing too many people.

For the same reason, some Investment Banks praise teamwork, notably in their structured finance department.

## Peter Grieve writes:

Not sure this is relevant, but I heard that snipers never ever make it to prisoner of war camps. When an identifiable someone has been deliberately targeting your friends as individuals, accidents always seem to happen in the prisoner-taking process.

# Markov Chains, from Nick White

Please correct me if I'm wrong, but am I right in thinking that the particular suitability of Markov models to interest rate derivs derives from the term structure of interest rates vis a vis the section of the wiki entry that states:

"Since the system changes randomly, it is generally impossible to predict the exact state of the system in the future. However, the statistical properties of the system at a great many steps in the future can often be described"

Here is an abstract of a paper that might be of interest.

We introduce a general class of interest rate models in which the value of pure discount bonds can be expressed as a functional of some (low-dimensional) Markov process. At the abstract level this class includes all current models of practical importance. By specifying these models in Markov-functional form, we obtain a specification which is efficient to implement. An additional advantage of Markov-functional models is the fact that the specification of the model can be such that the forward rate distribution implied by market option prices can be fitted exactly, which makes these models particularly suited for derivatives pricing. We give examples of Markov-functional models that are fitted to market prices of caps/floors and swaptions.

## Bruno Ombreux writes:

I just checked "Analysis of Financial Times Series", by Tsay. It is a relatively recent book so should be almost state of the art. It looks like Markov chains are used in two areas:

- better time series modeling, with regime switching.

- MCMC as a tool for Bayesian inference, whose main financialapplication at this point seems to be stochastic volatility models.

Practically, that would mean applications in "ever changing cycle" detection and option pricing/hedging.

On the first point, I personally toyed with Hidden Markov models. They work well in hindsight and are able to detect transitions between (high volatily / bear) and (low volatility / bull). But:

- this is all in hindsight. "Past performance is no guarantee…"
- this is low frequency data and I am not sure that the fact they could
detect the handful of past secular changes is that much useful. - a simple volatility threshold might do the job just as well without complication. - what you end up with is a probability transition matrix, which is not very helpful given that you are looking at only a few cycles.

… Maybe they could be more interesting in high frequency. I don't know.

On the second point, my opinion about option pricing is that the price of an OTC option is "as high as the buyer is prepared to pay". Models are an excuse. So I am not sure bayesian stochastic volatility models beat "seat of the pants" marketing.

If we look at the Niederhoffer-Osborne data on p. 899 we can see the number of times the market went from a don tick of various sizes to an uptick state. The reverse is also enumerated. I will take just the -1/8 state and +1/8 to illustrate the following. This matrix is the transition matrix by number of times each prior state led to the subsequent state.

-1/8       +1/8     total
-1/8      231        777      1008
+1/8      709        236       945

From this we can compute the probabilities of being in each state given the prior state in the previous time period (trade).

Probabilities

-1/8       +1/8      total
-1/8      .23        .77       1.00
+1/8      .75        .25       1.00

For Markov analysis we can make a prediction by multiplying the probability matrix by the vector which describes the current state. The result is a new vector of probabilities of being in the two states given the initial state. To predict two states ahead we multiply that by the probability matrix yet again. Let us take S as our starting state and P as our probability matrix. Then a prediction k steps ahead is given by:

S(k) = S(0) * P ^ k

Note that * is the matrix multiplication operator and ^k means to multiply P by itself k times.

What often happens is these matrices arrive at some steady state equilibrium after k iterations and we get the happy result that the probabilities are unchanged from iteration to iteration.

## Russ Sears writes:

the Society of Actuaries has many research articles using the Markov Chain process. To see them all type in "Markov chain" into their home web search. I believe one of the best papers to give and intuitive understanding of its use and power and limits, is Ms. Christiansen's paper in which she gives a brief intro to many different interest rate generators.

# An Idea for a Fundamental Theorem for the Productivity of Nations, from Bruno Ombreux

April 8, 2010 | 2 Comments

I think we have a problem that stems from a confusion between the asset side of the economy and the liability side of the economy. The assets are land, plants, people. They produce something. They create wealth. Sometimes this is called "the real economy."

The liabilities are financing the asset side, but they don't create wealth by themselves, or only at the margin (via tax arbitrage mainly). They are: stocks, bonds, etc… They are the oil in the engine, not the engine itself. The Fed is part of the liabilities.

What macroecomics is sorely lacking is a Modigliani-Miller theorem. As you know, the Modigiani-Miller theorem is a corporate finance finding that what matters is the asset side of the balance sheet, not the liabilities. The way it is financed doesn't change the value of a firm. It needs to be extended from corporate finance to macroeconomics, and I am happily providing it here as a conjecture, because at this stage it is not a theorem yet: "In a tax-free world, finance is irrelevant to the real economy." That means monetary policy is irrelevant, the Fed is irrelevant. What matters for wealth creation and growth are people, plants, and land.

Things were going well until the late 1990s because of the asset side:

1. reconstruction after WWII
2. baby boom
3. cheap oil
4. wave of innovation
5. no competition from emerging countries– they were communist, and, in hindsight, communism was great for Western Europe and the USA, because it meant less international competition

All these positive factors meant the asset side of the economy created huge value. It would have done so with or without monetary policy, with or without central banks, with or without banks actually. Now the positive factors are gone. The real economy is doing poorly and it will do so no matter what is done in terms of monetary policy, which is I repeat, irrelevant.

That's why we can have huge unemployment, that's the assets world, and a booming stock market, that's the liabilities world. They really operate independently.

## Alston Mabry writes:

I'll bite.

The last few years appear to have been a story of how much finance does matter, unless one argues that the global downturn was a purely secular matter coincident with a financial collapse. Now, I have felt that the importance of the cyclical downturn in real demand got less credit than it should have, but it would be tough to argue that finance doesn't matter.

At any given point, there is some interplay between the finance side and the real asset side. The nature of that interplay changes over different regimes. There are times when lax monetary policy is just "pushing on a string" because there is no demand waiting to be unleashed by loose money. There are other times when changes in policy can have a much larger effect. And monetary policy is just one vector of finance. Finance can matter a lot, in different ways, at different times.

The asset world and the liabilities world certainly can operate independently, but we may also be seeing the markets work as predictors of what is going to happen in the "real" world.

## Rocky Humbert disagrees:

I disagree that what macroeconomics is sorely lacking is a Modigliani-Miller theorem.

The practical interpretation of Modigliani-Miller is that leverage doesn't matter (to an enterprise) and while it has limited merit in structuring an enterprise during normal times, any sensible executive (or hedge fund manager) who has lived through a severe business contraction or credit squeeze will laugh at the notion that leverage doesn't matter. M-M correctly observes that a lousy business with 2% ROE is still a lousy business with 20% ROE (after 10:1 leverage). However, M-M doesn't say that a decent business with 7% ROE can be turned into a lousy business with 70% ROE (after 10:1 leverage). That is, leverage can't turn a lousy business into a great business, but it can turn a great business into a lousy business. After after that cyclical downturn, the company with less leverage will have less competition, more market share and greater unleveraged ROE. Lastly, if debt doesn't matter, why does anyone care about a rising national debt? Given the choice, I'd prefer a restoration of the Papal Vix Pervenit to a M-M in macroeconomics.

# Article on the Foundations of Calculus, from Nick White

Steven Strogatz has a great column in today's NYT (looks like the first in a series) about the foundations and intuition of calculus.

## Bruno Ombreux writes:

In the same vein, I read a nice book about the history of mathematics: Taming the Infinite. The link is to the hardcover edition, because it is such a beautiful book that it is better to buy it in hardcover.

I think it would be great if mathematics were taught together with their history. Before introducing a subject, the teacher would first explain the real world questions that led to its discovery, and talk about the people behind it. Some mathematicians led very interesting lives! This way, some maths would stop looking like they were pulled out of a hat and the kids would get interested.

## Pitt T. Maner III notes:

Note that Strogatz has received good reviews for his book The Calculus of Friendship. Interview with Alan Alda may be of interest too (aside–Alda should play Dr. Feynman in a movie before he gets too old, he has done it on stage in QED, it's an amazing resemblance especially his voice pattern).Dr. Strogatz discusses how his wonderful high school teacher reversed the roles and inspired him to "teach the teacher". The reverance and sincere praise for higher learning was an inspiration. Interesting overlap of life, emotions, and cold, hard math.

Strogatz discusses "balance theory" in a recent paper and wondered if it might have market implications? The plus and minus sign usage somewhat like Chair's. It also mentions Markov processes and different states:

The shifting of alliances and rivalries in a social group can be viewed as arising from an energy minimization process. For example, suppose you have two friends who happen to detest each other. The resulting awkwardness often resolves itself in one of two ways: either you drop one of your friends, or they find a way to reconcile. In such scenarios, the overall social stress corresponds to a kind of energy that relaxes over time as relationships switch from hostility to friendship or vice versa. This view, now known as balance theory, was first articulated by Heider [ fields ranging from anthropology to political science [1,2] and has since been applied in3,4].

# Negative Probabilities, from Rocky Humbert

March 17, 2010 | 1 Comment

Most of us view the probability of an event as being between zero and one. But this is a simplification. Negative probabilities exist in physics, and they "probably" exist in the markets.

Additionally, probabilities greater than 1 exist too. Probabilities which are less than zero or greater than one are called "extended probabilities."

This is the first paper that I've seen which builds a mathematical model of interest rate options for negative probabilities. Previous papers have dealt with "risk-neutral" and "psedo-probabilities." The authors also promise an upcoming paper that describes financial models for events with a probability. The paper makes good reading for those who enjoy dividing by zero, and taking the square root of a negative number.

For the less geeky, besides negative interest rates, can anyone think of some real world examples of negative probabilities? Or probabilities greater than one?Is "Hell freezing over" an extended probability?

Rocky Humbert, quantitative analyst, speculator and master chef, blogs as OneHonestMan.

If it is less than 0 or more than 1, then by definition, it is not a probability. It is not even a measure. They could call them anything, for instance "tiger-striped ferrraris", but they should not call them "negative probabilities".

The reason for a semantic discipline requirement, is that this tongue-in-cheek article, is targeted at finance people. People in the finance industry are generally clueless and take this kind of joke at face value.

Right now, I am studying Bayesian statistics, where they make ample use of calculation hacks and gimmicks. For instance they use Dirac masses as probability densities (height is infinite, width is zero, and area 1 ). But they know exactly what they are doing and nobody is fooled by the vocabulary.

That's different when the public is the banks or HF unwashed masses. For these, a dog should be called a dog, a cat a cat…

## Rocky Humbert responds:

While I have often have my tongue planted in cheek (as well as foot planted in mouth), that is actually not the case here.

Mr. Bruno's first sentence is entirely correct with respect to "classical" probability theory. However, he might consider the possibility that Extended Probability Theory is analogous to Einstein's Relativity Theory extending Classical Newtonian Physics. (i.e. there are practical applications of atomic physics in our mundane lives — one doesn't need to travel at the speed of light to see this.)

I'm not a mathematician (and I don't play one on TV either) but I'm told that negative probabilities have a long history for people thinking about the foundations of quantum theory. Feynman wrote about them and the concepts led to his initial work on quantum computing. (Basic quantum computers now exist.)

Additionally, in markets, I believe that "Dutch Books" may give rise to extended probabilities.

I can't quarrel that some folks in the finance industry are clueless. (C'est moi, Monsieur!) but I don't find Brownian Motion-based options pricing models entirely satisfying either… hence I try to keep an open mind.

## Jon Longtin writes:

I would caution against mixing mathematics with physics. Math is an (actually the only) absolute science, who's existence is defined completely in terms of stated rules and relationships. It is, at the end of the day, a very large body of definitions.

Probability, in the mathematical sense, is the chance that a particular outcome will happen, with the assumption that that outcome can at least happen. If an event can never happen its probably is zero, and if it always happen it is one. *Mathematically* to speak of events outside of this context is meaningless.

Physics is, well, physics. The world is the way it is and it's our job to describe it to the best of our ability. A tool that does this remarkably well is mathematics. Often, though, as we learn more the physical models have to be revised, expanded, and reinterpreted, given new information and insights. When we look at our new and improved models through the lens of the old model, thought, strange things happen. This is true with relatively, quantum mechanics, and when they discovered that the sun went around the earth, to name a few.

There are, for example, new materials that are characterized as having a negative index of refraction, n (a measure of how strongly materials bend light, and is the reason a pencil looks bent when in a glass of water); classically vacuum has n = 1 and air is about 1.0001 or so, water = 1.33, with values less than 1 physically impossible. There are, however, new materials being developed that do not naturally exist in nature, but rather are engineered structures that give the illusion of having a negative index. The point is no physics are being violated here; only that the model needs to be revised, and fitting the new material into the old model will result in surprising and sometimes counterintuitive understanding.

It is sometimes tempting to introduce an extension to the old model, such as e.g., negative refractive index and negative probability, but the more rigorous approach is to redefine the physical model from the ground up to capture the new phenomenon in a rigorous way.

Jon Longtin, Ph.D, is Associate Professor and Undergraduate Program Director, Department of Mechanical Engineering, State University of New York at Stony Brook

## Bruno Ombreux replies:

In the case of the negative probability article, they are quick to dismiss the obvious and parsimonious solution that everyone has been using, and replace it with some harebrained theory.

Negative interest rates are nothing new and not a problem. We have had plenty of trade-ables that have always been able to get negative, with active OTC option markets in them, eg crack spreads…

The simple solution is to use a normal law instead of log-normal, and if you are still not happy, to use the empirical distribution.

There is a fine volume recently out by the wonderfully polymathic Clifford Pickover called The Math Book. "From Pythagoras to the 57th Dimension, 250 Milestones in the History of Mathematics.

Reading each of these 250 entries is like doing a set of 25 push-ups for one's mind. And not just the left side, as the methodology behind fractal artwork indicates.

On the subject of squares of a negative number of which Rocky wrote, Dr. Pickover touches on the formerly ridiculed notion of imaginary numbers, the contributions of Bombelli, et al., and writes:

An imaginary number is one whose square has a negative value. The great mathematician Gottfried Leibniz called imaginary numbers 'a wonderful flight of God's Spirit; they are almost an amphibian between being and not being.' Because the square of any real number is positive, for centuries many mathematicians declared it impossible for a negative number to have a square root. Although various mathematicians had inklings of imaginary numbers, the history of imaginary numbers started to blossom in sixteenth-century Europe. The Italian engineer Rafael Bombelli, well known during his time for draining swamps, is today famous for his Algebra, published in 1572, that introduced a notation for √-1, which would be a valid solution for the equation x² + 1 = 0. He wrote, 'It was a wild thought in the judgment of many.' Numerous mathematicians were hesitant to 'believe' in imaginary numbers, including Descartes, who actually introduced the term imaginary as a kind of insult."

Leonard Euler in the eighteenth century introduced the symbol i for √-1 — for the first letter of the Latin word imaginarius — and we still use Euler's symbol today. Key advances in modern physics would not have been possible without the use of imaginary numbers, which ave aided physicists in a vast range of computations, including efficient calculations involving alternating currents, relativity theory, signal processing, fluid dynamics, and quantum mechanics. Imaginary numbers even play a role in the production of gorgeous fractal artworks that show a wealth of detail with increasing magnifications."From string theory to quantum theory, the deeper one studies physics, the closer one moves to pure mathematics. Some might even say that mathematics 'runs' reality in the same way that Microsoft's operating system runs a computer. Schrödinger's wave equation — which describes basic reality and events in terms of wave functions and probabilities — may be thought of as the evanescent substrate on which we all exist, and it relies on imaginary numbers.

Here is Dr. Pickover's website (well worth a look).

## Rocky Humbert replies:

With your reference to Dr. Pickover, you tied together many loose ends:

Mr. Maner alluded to the fact that there is a probability greater than one that "…vampires will invade the literary world and be a profitable genre." He referenced the epic drama: "Abraham Lincoln: Vampire Hunter " Low and behold, it was Dr. Pickover who invented vampire numbers– I would wager that this is the first time that negative probability, markets, Abraham Lincoln and vampires were all discussed in the same thread on Dailyspec. (The probability of this happening is comparable to the odds that the S&P will rise five more days in a row. I therefore conclude that this must be an omen, and I just bought ONE March S&P 116 call as a homage to negative probability and vampires.)

## Sushil Kedia writes:

The first and the simplest example of negative probability at work in the markets comes to my mind from the Chair's oft emphasis on deception in the markets.

Let me use an example:

A street conman's game very much prevalent in India, near the smaller train stations and ports, where the hustler holds a heavy bag of muslin with two hands and offers a wide peep inside for you to see a nicely mixed hoarde of coloured and natural peanuts. The odds offered are 3:1 for you to multiply your money if you lift up a coloured peanut. You rush in playing an unfair game apparently to your advantage. When you shove your hand in, the mouth of the bag is held much closer around your wrist than when you were inticed to take a game loaded in your favour.

You pull out the peanut and it is not coloured.

The trick deployed is that there is another bag within the bag containing only uncoloured peanuts.

In the awareness of the hustler, probability of the outcome is certain due to his ability at deception. In the awareness of the player the probability of the outcome is somewhere close to 50:50. In the awareness of an analyst like me who has burnt the hand that tried picking the coloured peanut many times 50/(50+50+100) or 0.25 assuming the hustler fails at closing out the bag with mixed peanuts forcing your hand into the hidden bag with only plain ones.

The difference in the probabilities known to the newbies and those who have burnt their hand and become analysts is the 0.75 gap which is really the negative probability on which the hustler is peddling his skill.

Replace the peanuts - plain and the mixed ones with earnings guidance and announcements and you realize the negative probability the masses face vis-a-vis the insiders assuming all else being equal.

Replace the peanuts - plain and the mixed ones with counted stats the pros are playing with and the bags of code (not only computational but simply deal flow informational) the glittering big firms can have.

So on and so forth.

With this perspective gaps in perception, information, imagination, awareness, model specification, ability to loot, peddle, hustle etc. etc. a concept of negative probability fits in well with comprehension. The Oracles of Delphi as explained in the Education of a Speculator played really on the negative probability the masses believe are non-existent and happy to live with such belief. Despair, disdain, pursuit of short-cuts, road to quick riches have all been built with bricks of negative probability.

The first and the simplest example of negative probability at work in the markets comes to my mind from the Chair's oft emphasis on deception in the markets.

Let me use an example:

A street conman's game very much prevalent in India, near the smaller train stations and ports, where the hustler holds a heavy bag of muslin with two hands and offers a wide peep inside for you to see a nicely mixed hoarde of coloured and natural peanuts. The odds offered are 3:1 for you to multiply your money if you lift up a coloured peanut. You rush in playing an unfair game apparently to your advantage. When you shove your hand in, the mouth of the bag is held much closer around your wrist than when you were inticed to take a game loaded in your favour.

You pull out the peanut and it is not coloured.

The trick deployed is that there is another bag within the bag containing only uncoloured peanuts.

In the awareness of the hustler, probability of the outcome is certain due to his ability at deception. In the awareness of the player the probability of the outcome is somewhere close to 50:50. In the awareness of an analyst like me who has burnt the hand that tried picking the coloured peanut many times 50/(50+50+100) or 0.25 assuming the hustler fails at closing out the bag with mixed peanuts forcing your hand into the hidden bag with only plain ones.

The difference in probabilities known to the newbies and the analysts is 0.5-0.25 = 0.25 the awareness advantage.

The difference in probabilities known to the analysts and the hustler is 1-0.25 = 0.75 the hustling advantage.

The difference in probabilities known to the hustler and the newbies is0.5-1.0 = -0.5 the ignorants negative probability

Awareness Advantage MINUS Ignorants negative probability = 0.25-(-0.5) = 0.75 = The Hustling Advantage or in other words, the negative probability is Awareness Advantage - Hustling Advantage.

Replace the peanuts - plain and the mixed ones with earnings guidance and announcements and you realize the negative probability the masses face vis-a-vis the insiders assuming all else being equal.

Replace the peanuts - plain and the mixed ones with counted stats the pros are playing with and the bags of code (not only computational but simply deal flow informational) the glittering big firms can have.

So on and so forth.

With this perspective gaps in perception, information, imagination, awareness, model specification, ability to loot, peddle, hustle etc. etc. a concept of negative probability fits in well with comprehension. The Oracles of Delphi as explained in the Education of a Speculator played really on the negative probability the masses believe are non-existent and happy to live with such belief. Despair, disdain, pursuit of short-cuts, road to quick riches have all been built with bricks of negative probability.

## T.K Marks writes:

On the subject of meals, I see today a poignant portrait of the food chain. These photos from Colorado are spectacular.

"…The starling seems to be completely unaware it is on the lunch menu as the bald eagle makes it attack at high speed…" On some level we've all been starlings at one point or another.

# U to the Tenth Power, by Victor Niederhoffer

March 13, 2010 | 4 Comments

It is interesting to speculate about what happens when the market is up 10 days in a row and then again at the open, since this has never happened before so one can't do it. This is a quandary for frequentists and perhaps mysteries of the Bacons, Dave and our readers could shed light, as I can't.

## Alston Mabry writes:

I'm looking at SPY closes, only one peak higher than this, back in Sep 1995, 12 up days in a row. lt's like standing on K2 and being able to see Everest from there. But maybe I'm just tossing coins…

I found a funny rule in Statistical Rules of Thumb, by Gerald van Belle, called the Rule of Threes: "Given no observed events in n trials, a 95% upper bound on the rate of occurrence is 3/n." There is one very simple demonstration based on the Poisson distribution.

The van Belle rule would say that in 1000 repetitions, it would be 19 to 1, that a decline would occur less than 300 occasions (i.e. a probability of 3/10), a truly precise but completely misleading answer in this case, especially when the underlying base estimate is 0.5. Sort of the way people talk about Microsoft's answer to a help question. However, in this case I predict a decline of 0.25, just to make the people waiting for a reversal crazy and even poorer.

# Bayesian Statistics, from Bruno Ombreux

February 6, 2010 | 4 Comments

So… I have studied the foundations of Statistics. Gone back to first principles. I am now convinced that the Bayesian approach is the way to go. It makes much more (common) sense than the rest.

The only reason Statistics today is not 100% Bayesian, is that the Reverend's paradigm has been made practical only recently, with the growth of computer power and the discovery of some really clever algorithms (Gibbs sampling etc…).

My idol, role-model, who I hold totally admire and who keeps me in a state of awe, Ronald Fisher, was a Bayesian. It is not obvious until you read his original articles, but he was. He was also a nice person in a time before computers, so he developed some simple not-entirely-Bayesian solutions to help mankind.

Bayesian statistics is the way to go… Only problem is that I have come to realize I know very little about them beyond basics.

My program for 2010 is to master Bayesian statistics. By "mastering" I mean being able to read a PhD thesis, but not necessarily to write one (I have a life).

There are many great statisticians on this list that could advise me on some books. I think Vic is a great Statistician, and Fisherian/Bayesian even if he doesn't realize it.

Bayesian Computation with R Albert
Bayesian Data Analysis Gelman & Carlin & Stern & Rubin
Bayesian Forecasting and Dynamic Models West & Harrison
Introduction to Modern Bayesian Econometrics Lancaster

What I need now is to know what are the best books in the field, and some form of advice on progression, the above books being a bit of a strange mix between obvious and difficult.

If you were a university teacher, what books would you require your students to read and in what order?

And should I read the original articles from the 1950s advocates, eg Savage, De Finetti or Jeffreys?

# International Bond Markets, from Andrea Ravano

January 30, 2010 | 1 Comment

Greek and Portuguese bonds are in a nasty spiral. Very little seems to be working in terms of convincing the markets to mop up some paper. Greece 3.7 2015 is now trading 86-86.5, yielding approx 6.6 pct and some long term Portugal bonds are down a point or so since yesterday. I don't think Europe is in any way capable of rescuing Greece, or anybody else for the matter; the virus will soon spread to Italy, as it suffers from the samle chronic high debt to gdp ratios as the afore mentioned countries. Thus the trade of the day could be long Bunds short Btps.

## Jeff Rollert writes:

Would it be unreasonable to compare the inability of any country to act as the world's military police, and in a similar sense, one country being the worlds bank?

Seems like the ECB built a wing on their house with wood full of termites.

I've always enjoyed the science fiction writers observation that the world will never unite until there is a non-Earth threat. Perhaps that includes monetary unions.

## Alston Mabry writes:

It used to be so simple: The Greeks would have a crisis, the drachma would fall, and the Neuro's would swarm down for sun and fun and economic stimulation. The Greeks then took the extra money and started another story on the house because they knew that keeping the cash was not a good long-term investment. You'd see half-finished buildings everywhere, bristling with rebar — just the local version of a savings account with a currency hedge.

Have you been to Athens recently? That's exactly what they have. Half-finished houses. They don't even bother covering the concrete. I was told that it was for tax reasons. As long as the house is unfinished, there are no real-estate taxes. So they don't finish their houses. This is very creative.

## Jim Sogi replies:

Same thing in Peru.

I didn't attend either event, but I remember in 2003 when Athens hosted the FISA Junior World Rowing Championships and then in 2004 Olympic Games someone made a comment about how clean everything was. It wasn't until about a week into Jr Worlds that someone finally noticed the grass on the sides of the highway between the athlete village and the rowing venue wasn't grass, but a green tarp covering heaps of trash.

The state of the art rowing venue is to my knowledge abandoned today. It was also only finished one week prior to Jr Worlds, and no one thought to anticipate the mid-August winds that sweep the city. The winds created such waves that the Men's eights heats had to jump ship and swim their boats between 500m and 1000m to cross the finish. Finals were reduced from 2000m to 1000m. The Games were lucky and didn't have this problem.

But what about selecting cities in order to build athletic facilities that will help the community in years to come? I wonder if there is been any research regarding future price performance of munis issued to build venues for Olympic Games. Most venues go unused after the event.

Another reason for the rebar sticking out the tops of buildings in some places is that they expect to build the building taller later, when money is available, but without a mechanism for collecting on debts there is little money available for lending, thus things tend to be paid for in cash, and built gradually. Here, with loans available, that strategy doesn't pay as well as borrowing the money to build a property to it's "best" economic use, as the cash flow is much worse on a partially built-on property - same land taxes, same land cost, lower return, higher hassle/permit costs for repeated small construction jobs.

# R. A. Fisher, from Bruno Ombreux

Otzi has no heirs [see Surprising Results Of Complete Mitochondrial Genome Of 5,000-Year-Old Mummy ] is not strictly proven, although they are right in a statistical sense. They can't say anything about Otzi, but they are right about the population he is issued from.

This combination of genetics and statistics gives me the opportunity to express my most complete admiration for R. A. Fisher.

Now that I dug a bit into the foundations of Statistics, I have come to realize that Fisher is one of the great geniuses of the 20th Century. Greater than Einstein in my opinion.

I don't understand why he is not more famous. Take for example, The Conditionality Principle. Fisher had the intuition and Birnbaum formalized it.

This thing is a moment of pure concentrated genius and joy.

I don't know anything else like that in science. In the sense that it is incredibly practical and down-to-earth while being at the same time very foundational and philosophical.

# Betting on Farm Land, from Jeff Watson

November 25, 2009 | 1 Comment

Here's a good article from Fortune/CNN-Money about the resilience of the agricultural real estate market. One never thinks much about agricultural real estate, but the value of the land reflects the amount a farmer can borrow to run his business and ability to put in his crops. Agricultural land has suffered far less in the recent downturn than residential and commercial real estate markets. Since 1945 the average size of a farm has increased from 213 to 461 acres, but yields have increased three to four fold. Still, the increased yields are working against the income of the farmer with each 1% increase in yield resulting in a 4.5% decrease in prices at the farm gate.

Despite this increase in size and productivity, the market in acreage is cognizant of the fact that we're losing an area of agricultural production the size of Ohio and Pennsylvania every 30 years, and our population is trending up. The bottom line is that the number of farms in the US is at an all time low, and that number is trending downward. The size of the farms are getting bigger, yields are going up, yet the arable amount of agriculture land is approaching a historical low basis the total population. We're down to 0.6 acres of arable farm land per person and that number is shrinking drastically.

The market is aware of these trends and the invisible hand is keeping the prices in agricultural land more in line with reality than the other real estate markets. Most farm land is sold without a broker and reporting of exact prices is difficult, as only the county assessors are privy to this most sensitive of information. Still, the broad trend of prices is up, and many are placing huge bets on the rise. Farmers are mostly close-mouthed, and asking a farmer how many acres he owns or how many head of cattle he has is tantamount to asking him how much money he has in the bank. Simply not good manners.

Jeff Watson, surfer, speculator and art connoisseur, blogs as MasterOfTheUniverse.

## Scott Brooks respectfully disagrees:

I have to respectfully disagree with much of what Jeff has written here, though. He is right that the the farming landscape (both figuratively and literally) has changed immensely and keeps changing over time.

But as to farmers not wanting people to know how much land or how much cattle or how much crops they're running, that may be true, but it's certainly not like how much money they have in the bank. You're money in the bank can be kept a secret, but the amount of land you have is pretty much wide open for the world to see. I can go to the FSA office and use their computer system to draw boundaries around a person's property and it will tell me how many acres they have….plus, anyone who lives in the area for any period of time knows how much land farmer Jones has. As to how many heads of cattle one runs, that's open for the world to see, too. And on top of that, if you ask a rancher how many head of cattle they are running, it's been my experience that they will usually tell you.

Most farmers that I know do not have to borrow against their land to run their operations. The FSA office, the local grain elevator, or the local farmers bank will almost always loan money to farmers to grow their crops. I know some pretty awful farmers and they still get the inputs fronted to them by these entities.

Farm land pricing is very regional in nature and depends on the type of land. Bottom land in the farming country is very valuable in terms of income. Hay ground, pasture and non-bottom crop ground is probably next in pecking order. Land in my area of MO, is priced low compared to land two counties over and much lower than land two counties north in Iowa. Scrub land, the land that was of little or no value a few years ago, is now very valuable as hunters (like myself) come in and want to buy up property for recreation.

The little farmer with 461 acres is not what you think he is. Take me and my 522 acres. Based on the standards of the study cited, I am above average in terms of farm size…..but…..I do NOT farm my land (I don't even live on my farm). So what this changes is perspective. I may not farm my land, but it does get farmed. What happens is a farmer (who owns around 750 acres within their family) rents my land for farming purposes (bottom land for crops, grasslands for hay, and other lands for spreading excess manure from his pig operation).

You see, a farmer usually can not make much of a living off 461 acres. So real farmers (non-corporate farmers) rent up a bunch of land to farm. My farmer rents my land and a bunch of other land so that he has agricultural control over ~ 3000 - 3500 acres of land (3,200 acres is the equivalent of 5 square miles of land……640 acres is equal to 1 square mile). The family can run crops, do contour farming, hill top farming, cut hay, run cattle, build hog barns, and many other activities in order to make money.

As to the pricing of crops, I don't believe that we'll actually know the value of the crop (grain or cattle) as long as the government keeps subsidizing their values.

The discussion of subsidies is a whole other post that I don't have time to go into, but suffice it to say, a book could be written on the subject (and several probably have). I'll just say this final thought about them. It is my opinion that the government is like a drug pusher, that once they get their hooks in you with the drug of subsidies, you can't get off them.

As to farm land prices being in line with reality goes……….. This is just false. My farm has quadrupled in value since I bought it 10 years ago, and I've had offers to purchase it for 7 times what I paid for as recently as 3 years ago.

Farm prices skyrocketed in the boom, but they also have not suffered as much as other real estate in the downturn. Most farms in my area, quadrupled in value over the last ten years. They actually quadrupled in 7 years and have remained fairly steady over the last 3. My farm was an exception to the rule due to the fact that I ran a Quality Deer Management Program on my farm and had 7 years of records and pictures of our progress. Due to the work I did, wealthy hunters from out of state were inquiring about purchasing my land for hunting purposes. Those offers have mostly dried up due to the economic downturn.

Obviously there is more to write on farming, but I have a 4 hour and 15 minute drive home from my farm (been up here hunting for the last two weeks) back to my home in St. Louis, so I'm gonna cut it off here.

But I will say this before I sign off. Everything that I have said is very much regionally based. Remember, farming is VERY regional. So what I've written here may not be germane to what is happening in northern Iowa.

So with that being said, what Jeff Watson has written may be true overall, but it's not in my area……so take what I've written with the appropriate grain of salt.

Here in France a state agency called SAFER records land sales. Their database holds more than 6 million transactions starting in 1970. They have less exhaustive data from 1950.

Unfortunately their database is on a pay-per-use access. But browsing it one can conclude:

- as Scott said, prices are very location dependent. There are literally thousands of micro-markets. Poor grassland in remote locations can be acquired for less than 1000 euros/hectare. Some wine land in Burgundy can fetch several million euros/hectare. (1 hectare = 2.47 acres).

- when one looks at the average of all transactions, land is a very good inflation hedge.

In real terms, prices today are a bit higher than in 1950. Land preserved its value through the inflationary 60s and 70s!

Actually, it looks like land is a leveraged inflation bet, with asymmetric payoffs around a CPI threshold. Above the threshold, one gain more than inflation, below the threshold, one loses in real terms.

On this long-term chart, there are 3 periods:

- 1950-1979: land prices gained in real terms. They increased faster than inflation. - 1980-1995: land prices decreased in real terms, while there was low but still positive inflation. - 1996-present: prices are gaining again in real term, by more than official CPI.

The settlers in Plymouth had experience in growing the cereals that were the major foodstuffs for Europeans in the 17th century: Barley, Oats, Rye, and Wheat. It strikes me as highly improbable that the one edible grass born in the Americas - corn (maize) - would have been so surprising to the Plymouth colonists that, but for the help of the Wampanoags, they would not have known what to do with it. The colonists called it "Indian corn" because it was so different from the types of corns (or grains) they were used to growing and eating; but the name itself suggests that they hardly needed instruction in planting (their colony was named The Plimouth Plantation).

Here is Bradford's version of how "Indian corn" was discovered:

"They also found two of the Indian's houses covered with mats, and some of their implements in them; but the people had run away and could not be seen. They also found more corn, and beans of various colours. These they brought away, intending to give them full satisfaction (repayment) when they should meet with any of them, - as about six months afterwards they did… And it is to be noted as a special providence of God, and a great mercy to this poor people, that they thus got seed to plant corn the next year, or they might have starved; for they had none, nor any likelihood of getting any, till too late for the planting season."

# Igon Values, from Victor Niederhoffer

November 23, 2009 | 2 Comments

My second-oldest daughter Katie's post on Igon Values is very relevant to our field. Who are the useful idiots in our field and how can their self-serving posings be used for profit? Let us start in the Midwest. Or the islands.

I thought the "igon values" thing was a joke, but upon reading Pinker's review of Gladwell, I find it isn't. For one who writes about research, as Gladwell does, to not be able to sit down and work linear algebra problems is fine (me neither), but to not know that there is a word, "eigenvalue", which may arise in conversation with scientists — that's embarrassing.

The issue of "igon values" bears on the more general issue of knowledge production and dissemination. I like watching science shows, but even when I'm watching a high-quality show like NOVA, I'm wondering, "Where are the folks, with the same level of expertise, who think at least some of this is crap? What is their critique?" It often seems that a well-communicated disagreement can help the audience understand more of what's important. And actually, a good example is Pinker's review of Gladwell.

In most or all fields of research, one could assemble a group of experts who have similar training and knowledge, but who disagree on important points in the field, especially at the boundaries of discovery. Then you have popular writers trying to understand and condense some field into book form — but if the experts don't agree, how can the popularizers possibly be "right"?

## Stefan Jovanovich writes:

Mr. Mabry is far too kind. Scientists talk their books all the time. A scientific reputation is made by presenting a theory that is striking, original and difficult to evaluate.  That theory becomes the scientist's brand. His/her future in academia is tied to the success of that brand.  Few, if any scientists, are crazy/honest/selfless enough to challenge the truth of their brand.  Hence, Heisenberg's comment that "the progress of science can be measured by professors' funerals."

In science, practitioners would rarely be lying about what they are doing. But in markets, who tells the "truth," unless that truth is consistent with the talker's book? Then there is the sheer volume of new "knowledge" produced on a daily basis. How does one cope?

## Bruno Ombreux writes:

Here is a suggestion. I believe it is linked to the Igon.

I am halfway through a book that is dealing with all these issues: Scientific Method in Practice by Hugh G. Gauch. It sheds light and fosters reflections on such things as scientific questions and methods, disproving or proving hypotheses…

It is a book about science. Since good trading is a science, this is a book about good trading too. Good trading is necessarily scientific, because good trading requires good predictions. Only science can yield good predictions. If trading is not scientific, it can't be good.

This is also a philosophical book. After a few chapters, I have enough philosophical ammunition to completely destroy the Black Swan school, on epistemological grounds. The Black Swan ideas that we cannot have models that work, that variance is either infinite or undetermined, are just as naive, and far less nuanced, version of David Hume's radical skepticism. In one sentence: we can't know anything. Scientific Method in Practice advises not to waste time arguing with radical skeptics. They are not targeting science, but common sense. Common sense is literally what humans can sense in common. In this case, we all can measure variances. Common sense is a key presupposition of science. Without common sense, there can be no science. Without science, there cannot be any debate between scientists and radical skeptics, since the later are saying in effect that the former don't exist.

Incidentally, the fat tails debate wonderfully illustrates one problem mentioned in the book, that is the underdetermination of theory by data. Observing fat tails, I can find offhand a bunch of explanations:

1. power law
2. slowly converging normal
3. Student
4. truncated Levy flight
5. mixture
6. Markov switching model
7. Agent-based dynamics

The same evidence produces a handful of theories. We are confronted to the issue of theory choice. In this case, I would start by getting rid of those that don't make predictions. Power laws would be the first casualty.

Now, on to another book recommendation: a first-year course in linear algebra and as such is related to the original topic. "Igon values for dummies," if you want. And it is free. This is a useful book for traders, because it is impossible to understand any recent article on economics or statistics without at least a passing knowledge of linear algebra.

I don't think mastery of Igon values is required to trade well, but other concepts can be very useful. For instance, the notion of projections, covered in chapter 3.VI, really helped me understand multicollinearity in regressions. Multicollinearity is the rule in financial time series. Often, its presence is not a problem, but you'd better know about it and when it can be a problem.

Combining this book's chapter 3.VI.2 about Gram-Schmidt Orthogonalization, with chapter 3.2.3 of this other free book, one gets a clear understanding of multicollinearity.

## Jack Tierney writes:

Being unfamiliar with eigenvalues (whether spelled correctly or not) led me to follow the threads in Katie's article. Those threads, in turn, led to still others. I finally landed on this.

The author laments the increasing propensity of Rhodes Scholars to go into the world of finance as opposed to some of the nobler scientific fields that once claimed so many of those blessed by old Cecil's beneficence.

"This break in an almost century-old pattern coincided with great increases in occupational earnings differentials, which have continued to grow, seemingly exponentially…the differentials in earnings…were often rationalized by Rhodes scholars as reasonable additional compensation to balance the lower standing of business jobs among their peers. "When differentials could become a hundredfold or far more — and as investment banking and similar firms started to recruit young Rhodes scholars who had degrees in math, physics or even history, English and theology — the yawning prospective wealth chasm understandably became impossible for many to ignore…"

So there we have it. Offer enough money and even the brightest will sell out. Let a dilettante like Gladwell emulate them, though, and the wrath of the informed will be merciless (just follow some of the threads and you'll discover that Kate's handling of Gladwell was relatively humane).

However, numerous responses seemed refer to the incalculable worth of the scientific method and were it adhered to, we would all be much better off and far less likely to be exposed to the ditherings of Gladwell et al.

Back in '93 a remarkable book written by a woman embittered by her brother's courtroom experiences hit the best seller list. It was "Whores of the Court" and detailed the lengths to which those supposedly trained in the scientific method quite easily (and lucratively) sold their conclusions. Each side could present "experts" with similarly impressive credentials; each side had access to the same evidentiary material; yet their conclusions could not
have been more different.

It might be legitimately argued that psychiatrists/psychologists aren't scientists in the pure sense of the word. Currently, however, we have scientists whose credentials most definitely measure up. Yet on issues ranging from the efficacy of ethanol to global warming to the amount of oil left within the earth's crust, their conclusions couldn't be more disparate. To put it bluntly, our scientists' opinions are for sale and this is occurring as government policy is
more and more determined by their conclusions.

Whose opinions are the most sought after and well rewarded (at least through speaking engagements, articles in the mainstream journals, and in research grants)? Generally, those whose views are the most dire or the least apocalyptic. This, in itself, is a sad development. But increasingly scientists whose expertise lay elsewhere are chiming in on one side or the other. As a result we are faced with promotions that announce that "X Number of PhDs Support Global Warming Theory", or "Y Number of PhDs Claim Peak Oil is a Sham."

I am increasingly exposed to individuals who claim (and firmly believe) that their opinion is as good as anyone else's, that it's unnecessary to study both sides of an issue, that it is quite OK to shout down a speaker whose views diverge from yours, and that it's quite alright to do whatever it takes to get whatever it is one wants.

In such a world, is Gladwell to be condemned or lauded? Are the newly minted Rhodes Scholars so misguided in pursuing wealth? Are scientists who missed the gravy train to be faulted for making a last mad dash for the gold ring on the caboose? Was Linus Pauling correct in observing that peers are nothing more than people who pee together?

This post reminded me of the book "Psychology of Intelligence Analysis" which contains these guidelines:

"Start out by making certain you are asking–or being asked–the right questions."

"Relying only on information that is automatically delivered to you will probably not solve all your analytical problems."

"Do not be misled by the fact that so much evidence supports your preconceived idea of which is the most likely hypothesis. That same evidence may be consistent with several different hypotheses."

"Proceed by trying to reject hypotheses rather than confirm them. The most likely hypothesis is usually the one with the least evidence against it, not the one with the most evidence for it."

## Chris Tucker replies:

Wow.  Some great reading in that book.  Thanks, Al. This is from Chap. 2 "Why Can't We See What Is There To Be Seen?":

People tend to think of perception as a passive process. We see, hear, smell, taste or feel stimuli that impinge upon our senses. We think that if we are at all objective, we record what is actually there. Yet perception is demonstrably an active rather than a passive process; it constructs rather than records "reality." Perception implies understanding as well as awareness. It is a process of inference in which people construct their own version of reality on the basis of information provided through the five senses.

This is so important in my line of air traffic control.  I am constantly telling trainees that listening is not something that happens to them, it is something one must actively engage in.  Upon hearing a pilot read back a clearance, whether it be an altitude, heading, speed or route, one must pay close attention to what is said and to check it against what is expected to be heard.  It is common for trainees to simply assume that they heard the correct readback and disconcerting to them when it is pointed out that this was not the case.  We spend a great deal of time teaching them how to listen attentively.

Another facet that I have mentioned before is teaching them to get the data from the scope — to look at groundspeeds and recognize overtakes, to look at altitudes and calculate rates of climb or descent, to look at aircraft types and make hypotheses about expected performance, to look at routes and destinations and see who has to get below whom, and to create plans based on all of these.  And then to observe and check hypotheses, again and again to make sure that what one expected to happen is really happening.  And if not, how to take steps to create the reality one intends.

The key to improvement in these areas is a combination of repeated exposure and active thinking about the available data. Exposure alone can make some tasks become automatic, but active thinking and attentiveness can accelerate learning and skill acquisition.

Gladwell self styles as a translator from the arcane indecipherable world of science to the everyday world of business and laymen. A good translator must understand the vocabulary of the original source language and must have a command of the vocabulary of the target language. However a command of the two relevant vocabularies is not sufficient. If it were computers would be the best translators.

What Gladwell lacks is semantic comprehension. It is often not sufficient to merely translate the words without a deeper understanding of the content. His Igon Value mistake is a glaring example.

Clearly his substitution of Igon Value for eigenvalue comes from only hearing the word as opposed to actually reading it in a book. Perhaps someone explained it in a phone or lunch conversation and Gladwell seized on it as an interesting buzz word.

Eigenvalues are actually a very beautiful construct in linear algebra. A simple intuitive way to look at them is the amount by which a quantity is stretched in a certain dimension. Suppose a stock or mutual fund has a beta of 2 and an alpha of zero. The equation is:

stock return = 2 * market return + zero

The eigenvalue for the above system is simply 2 because it stretches the market return by a factor of two.

The idea generalizes to 2 or more dimensions. Each dimension of a linear system has its own eigenvalue. If you have ever looked at yourself in a fun house mirror then you can understand this idea. The mirror that makes you look tall and thin has a stretching eigenvalue in the vertical direction and a shrinking eigenvalue (<1) in the horizontal direction.

Like the fun house mirror a matrix can be thought of as a transformation or mapping of one image to another. If one takes the eigenvalues of a matrix and multiplies them together the product acts much like a volume just as length times width of a rectangle gives the area. In Linear Algebra this volume is called the determinant. If any of the eigenvalues is zero then one of the dimensions has collapsed. It also means the determinant will be zero, the system of equations cannot be solved and any regression will be meaningless.

I have never seen any financial model take into account a determinant. Yet there seems to be a grudging acceptance of the idea that when a financial panic hits all the correlations approach 1 as people seek liquidity by whatever means necessary. Rather than simply look at risk from a simple beta model or VAR approach perhaps the proper way to model disaster is the determinant where all the risk are multiplied together.

Dr. McDonnell is the author of Optimal Portfolio Modeling, Wiley, 2008

# Deception, from Victor Niederhoffer

The best books on deception are the best books about the market. The best book about the market according to Martin Shubik is Ben Green's Horse Trading. I would add that there is a good section on deception in EdSpec. And I would point out that a systematic categorization of deception is essential and this is available in the ecology literature following J.R. Krebs's citations on deception in various species, especially monkeys.

Of course, as I've eulogized no end, The Farming Game by Bryan Jones also has much to say on deception in the buying and selling of livestock, and does so with wit and insight.

## Russ Herrold re: The Farming Game:

I purchased The Farming Game on your recommendation and enjoyed it.  It was a bit dated as to price examples (they look like a series from the mid 1960s to the mid 1970s), but the underlying principles remain sound. The book starts a bit slowly, setting up some stereotype character sketches, and then strings them together a bit, a bit later in the book.

## Kim Zussman writes:

Here is my Deception reading list:

Stocks for the Long Run, Siegel
Irrational Exuberance, Shiller
A Random Walk Down Wall St., Malkiel (efficient markets)
Beating the Street, Lynch (inefficient markets)
Trade Like a Hedge Fund: 20 Successful Uncorrelated Strategies & Techniques to Winning Profits, Altucher
The Intelligent Investor, Benjamin Graham (value)
Common Stocks and Uncommon Profits and Other Writings, Fisher (growth)
Futures: Fundamental Analysis, Schwager
Technical Analysis of the Financial Markets: A Comprehensive Guide to Trading Methods and Applications, Murphy
Contrarian Investment Strategies - The Next Generation
, Dreyman
Trend Following: How Great Traders Make Millions in Up or Down Markets, Covel
Momentum Stock Selection: Using The Momentum Method For Maximum Profits, Bernstein

(eigenvector = deception)

To Dr. Zussman's excellent list, I would add another one very much off the radar screen. In Hostile Territory by Gerald Westerby is purported to be written by a former Mossad agent and profiles his adventures in Africa, the ME and Europe.  I consider it of the best books written on the manipulation of human perceptions, mental flaws and frailties.  I try to read it once a year to condition myself to avoid the traps. It is right up there with Cialdini, but the dynamic and life threatening challenges faced by the author are much more entertaining while providing extraordinary lessons on the subtleties of behavior.

I found the walk through on structuring a diversified [here: agri-]business very approachable, and anticipate lending it out to give context for further discussions to some I work with and mentor.

The best book I ever read on deception was called The Game by Neil Strauss. This book is the holy grail for pickup artists, but the lessons easily translate into all areas of life from sports to trading to games. It was very entertaining and well written, znd Strauss gives point by point instructions on how to manipulate, deceive, obfuscate, hypnotize, and control your opponent or object of desire. Strauss takes time to delve into the science of how to pick up women, and believes in rigorous testing and the book surprisingly isn't as misogynistic as one would expect.

## Bruno Ombreux writes:

One absolute classic is Arthur Schopenhauer's The Art of Controversy. It also goes by a different title: "The Art of Being Right". Here is a Wikipedia article with the full list of stratagems. And it is available for free at Gutenberg.

## Kim Zussman writes:

The most respected investment books of the 20th century all have eigenproblem of hidden utility. Even when authors are intellectually honest, it's hard to understand how they could escape distortion induced by rewards.

Some are selling their strategy (read my book but invest with me), talking their book (I'm deep into growth or value, so please buy these), pandering the academy (status as published professor), making a career of teaching how to trade, increasing status, creating a legacy, etc. This is similar to the more general, "how many friends do you have who don't profit from you?"

## Bruno Ombreux responds:

I haven't read all the books in Dr. Zussman's list, but among those I've read, I think two are not deceptions:

A Random Walk Down Wall St., Malkiel (efficient markets)

Most investors would be better off reading this book and stopping there. Also:

The Intelligent Investor, Benjamin Graham (value)

I haven't finished this book because after the first two chapters I realized it was just a watered down version of the first edition of Security Analysis, from the same author + Dodd.

Security Analysis is an excellent book that makes excellent points for the era it was written in. Their technique of looking into detail at companies accounts is similar to detective work, which itself is an application of the scientific method. In my opinion, this kind of financial analysis is a valid way to proceed.

The nature of deception may be much deeper than many authors make out. I would say that the origin of all deception is in fact self-deception and that the supposed 'deceiver' is doing nothing more than moving into the vacant space within our understanding.

## George Parkanyi writes:

There is a saying.  "Fool me once, shame on you.  Fool me twice, shame on me."  To me it's just a given that traders, particularly those trading in size, use techniques to mask their intentions.  And sure, those that have knowledge of them, run stops.  That's just one of many influences that make financial instruments wiggle on a day-to-day basis, and you would not only have to sort out what is "deceptive" behaviour vs stupid vs herd behaviour, but whether the deception was or was not in your favour.  Unless you have a large network of people you can call on the inside that can give you information that helps you take the temperature of a given market, I don't see the point of trying to personify this market move or that market move as "deception", especially in a big liquid market that is essentially a non-linear system subject to multiple influences. If there's a pattern that you detect and can exploit then so be it.  But does it matter if it is "deception" vs. sentiment or just a big whale moving through?

Don't get me wrong. Reading about deception is certainly interesting. As a Scout leader, Arthur Baden-Powell's role in the Battle of Mafeking during the Boer War is an excellent example.  In fact, BP's entire early career was based on deception.  But I personally don't see the value in getting overly concerned about deception in the markets, though I understand that others do.

I think if you have a general sense of the day-to-day character of a market that you have researched and trade regularly, and do some research to try to anticipate macro influences on that market that might cause it to trend, the rest can be handled with money management.

## Stefan Jovanovich replies:

Baden Powell's energy as a commander was probably the decisive factor in having the deception succeed:

From British Battles:

Baden-Powell conducted the defence of the town with great energy and resource, leading the Boers to believe there was a larger garrison than was the case. In November 1899 Baden-Powell launched a series of raids on the Boers lines that caused him some casualties but made the Boers wary of the garrison.

Initially the Mafeking garrison had no artillery. Baden-Powell improvised various items to look like real guns and trains, while engineers manufactured a gun, known as the "Wolf", from a length of steel pipe. The Boers used the 2 two inch guns they had captured from Dr Jamieson to bombard the town. Dud shells fired from these guns were reworked and discharged at the Boer lines from the Wolf. An officer found an old muzzle loading naval gun serving as a gate post. This gun was christened "Lord Nelson" and drafted into service. Dynamite grenades were manufactured and thrown at the Boer lines and a small railway line was built across the town.

In sharp contrast to the indolent Ladysmith garrison, Baden-Powell kept his men constantly on the move, raiding the Boer lines and keeping the besiegers on their toes.

Atlas Shrugged not only speaks of deception, but the deceivers are open about their deception. The deceivers/looters are like gangsters who are in complete control in kick sand in the faces of the producers, daring them to say A is A and damning them if they do, all the while fooling the masses with their A is B pablum. The parallels to our world today are stunning.

# Applied Longitudinal Data Analysis, reviewed by Bruno Ombreux

"Applied Longitudinal Data Analysis" by Judith Singer and John Willett is simply the best book on longitudinal studies I have read.

It is pedagogical in the extreme, clearly written with the intention to teach something practical to the reader. There is very little math, and many real world examples. The authors even manage to explain maximum likelihood estimation in plain English, without formulas!

Actually, sometimes it is so didactic, so simple, that I couldn't help thinking "come on, I am not that stupid," and found myself wishing the authors were a bit faster and less detailed. But I'd rather have it this way, than some other books that read like the author is writing for aspiring Field Medalists. And clearly, the authors have some classroom experience. Where they are slow and repetitive must be where they found their students had difficulties.

So overall, an almost perfect book and I would be happy if all statistics books were like this one.

It is missing only three things:

- A chapter about combining longitudinal analysis with advanced time series methods, for instance what happens when the individual time series are GARCH or something. This book is oriented toward medicine or the social sciences, so it is lacking a bit on time series.

- Resampling methods.

- Some R examples. The book mentions almost every statistical software under the sun, except R.

But except for the R aspect, the other subjects are perhaps a bit too advanced and could be the object of a second book. In its present state, this book can be read by someone with very little math and statistics background. Some knowledge of regression is enough to start with, and the reader walks out with the capability to perform real-world longitudinal analysis. Adding advanced stuff would be for a second book, which I hope they'll write.

# TA Patterns, from James Sogi

October 23, 2009 | 2 Comments

Now that statistical quant is being denigrated on the Street, I am seeing these old TA John Magee patterns that haven't worked for years appearing. Like broadening top. Anyone else?

One of the weaknesses of statistics is the problem of changing cycles, and the appearance of hidden or unknown variables in a data series. Though it is prospective looking, it still uses historical data. The other issue is the role of randomness where very very long runs can appear, even in random series.

## Bruno Ombreux defends statistics:

I don't think statistics are weak, but the present environment is unique. For decades, we enjoyed a relatively stable environment and a capitalistic system. This has changed. All bets are off.

The problem is that the type of environment we are in, for instance consumer deleveraging while the Fed is money printing, has not happened very often. There are some similar situations in the 1950s, and some in the 1930s. That's not a lot. Therefore we can't even condition on the environment, because we don't have enough data for such environment.

I had come to rely on statistics. For the past few months, nothing has been working for me. I am not having a good year. Actually, this is my worst year since 1999 (compared to the index) and since 2002 (in absolute terms).

# Le SuccÃ¨s du MacDo, from Pitt Maner III

October 5, 2009 | 1 Comment

It is surprising to hear from the Daily Telegraph that business at traditional brasseries and bistros is in a freefall in France. Is it related to current economics, taxation of small business, red tape/licenses associated with owning small business , rent, convenience, demographic changes, Internet/advertising influences… what gives?

However, even if there were a last-minute u-turn at the Louvre, statistics suggest the battle of Le Big Macs has already been lost. France has become McDonald's biggest market in the world outside of the US, according to the chain. While business in traditional brasseries and bistros is in freefall, the fast food group opened 30 new outlets last year in France and welcomed 450 million customers, up 11 per cent on the previous year.

My recollection in 1984 in Sweden and Norway was that McDonald's used slightly different sauces on the Big Mac to appeal to local tastes but not sure if that is done these days and if regional tastes are considered.

I don't have hard numbers, but every day I walk past tens of Parisian food outlets; here are my observations:

- What is specifically declining are brasseries and bistros in their use for the noon meal.

- Cafés and restaurants, when they are well run, are still thriving. The terraces and interiors of those in my section are always full.

- McDonald's has been a huge success, particularly with teenagers. That's nothing new; it has been going on for decades.

My tentative explanations:

- Changing workers' eating habits. Fewer long noon lunches, more take-away and fast food.

- Traditional brasseries/bistros are too old-fashioned. They don't look modern inside so don't appeal to younger people.

- Cost of operating a brasserie/bistro: mainly staff, but also rents in Paris.

- Take-away food was cheaper. It was taxed at 5.5% VAT whereas non-take away food was taxed at 19.6%. But that changed a couple months ago; now VAT is 5.5% for everybody.

To answer your second remark, regional tastes are considered. The food at a French McDonald's is different from the food at a US McDonald's. The meat is of much higher quality and there is a choice of salads. The guy who spearheaded local variation in France did it which so much success that he is now the head of McDonald's Europe.

A cut in VAT tax rate, that is interesting. What kind of "consumer's strike" do we need in the U.S. for a similar capitulation to occur?

## Bruno Ombreux replies:

This is a bit complicated.

This idea of cutting VAT for restaurants predates the crisis. It was first tried unsuccessfully by Chirac. It was also a Sarkozy campaign promise.

But France can't cut VAT without an agreement from the EU. I am not sure why, but it must be because part of the VAT is funding the EU. At some point, we had a Polish veto [and a German one]. The Polish president didn't want France to cut VAT! Sarkozy must have threatened him with nuclear strikes and now France can decrease VAT.

I am plenty pleased. The cost of my morning coffee has gone down 12.5% (the café manager has kept part of the tax rebate for himself!)

# Look at The Romans, from Jeff Watson

The entire Roman Empire could be described as an economic stimulus package run amok. It conquered and grew with each conquest bringing in more booty of land, silver, gold and free labor (Slaves). And each conquest led to more conquests, and on and on ad infinitum . But the net effect of this plunder, for which no productive work was done, was only the need for more stimulus. It did not stimulate real prosperity, but instead undermined prosperity, much like the Spanish were to realize 1500 years later. First, slaves bought by rich farmers undermined the free labor market and ruined many small farmers who weren't able to compete. And then, imported wheat from the new provinces and conquered areas, caused the wheat market to collapse and affected the larger scale farmers.. Rome was by then dependent on foreign sources for its food, being a net importer for the first time by the first century AD.. In the first century AD, Roman conquests reached the tipping point of diminishing returns and the first physical effects of the downward spiral were felt. However, borders and trade routes still had to be protected, armies fed and clothed, the bureaucracy paid, pet projects built, the increasing graft of public officials, and the ever increasing demand by the public for bread and the circus. Therefore, the money had to come from somewhere, and that somewhere was out of the pockets of the productive, and also an increase of the money supply. Caesar Augustus tried to increase the money supply by having the slaves work around the clock in the silver mines to produce more silver. More silver was produced and much more coinage was seen on the street and in the vaults, but the net effect of this was inflation. Still, the leading economists and leaders of that era still were bent on fiddling with the economy and this led to Nero reducing the silver content of the coinage and a recall of money with higher silver content so it could be melted down and alloyed, causing more inflation and currency debasement. The Roman Treasury even tried a direct cash stimulus to the citizens, releasing over three million pieces of silver on to the street to kick start a depressed economy, but the result was inflation.

The Edict of Diocletian in 301AD was one of the bigger blunders in the Roman Empire. In order to control the spiraling inflation, in no small part caused by his tax and spend policies, Diocletian, along with his co-emperor and fellow autocrat, Maximian, issued an edict that mandated against selling at prices above a certain level, for an entire list of staple items, with the mandatory penalty of death for violation of the edict. The law of unintended consequences caused merchants to hold back product rather than sell at a loss, and another edict against hoarding was issued, again violations of this law carrying the death penalty. Facing many business failures and a major recession, Diocletian took the ultimate step. He arbitrarily issued another edict that made it punishable by death for one to go out of business (something that would have been the end result in Atlas Shrugged had the looters prevailed), and that each son had to work at his father's occupation. Although many university scholars like to present Diocletian as a reformer, visionary, vessel of change, and good leader who was able to hold everything together, he tripled the number of laws and regulations and greatly expanded government's rule over the daily lives of the population. Diocletian was a personally greedy petty despot, making Boss Tweed look like Mother Theresa, because he had absolute authority over an entire empire. His complex holdings necessitated the need for two finance ministers, one for the empire, and one to manage his personal portfolio. He went on a hiring binge of government workers, increasing the number of lifetime civil servants in Rome from 15,000 to 60,000, although that exact number is in dispute. Along with the lifetime civil servants was a corresponding increase in clerks, scribes, accountants, and especially tax collectors. He increased the bureaucracy by dividing up states, thereby duplicating local bureaucracies on a large scale. The early Christian activist and author Lactantius claimed that there were now more men in Rome using tax money than were paying it, a condition which is starting to reappear in today's increasingly socialistic societies. The whole scene during Diocletian was reminiscent of Terry Gillam's excellent movie, Brazil. Ultimately for reasons that are in dispute among scholars, Diocletian ended up abdicating shortly after his 20th year run in power.

Modern academics tend to treat the fall of Rome as an inevitable consequence of a civilization run amok and discretely blame free market forces for the demise. They fail to recognise the impact that rules, regulations, debt, high taxes, and welfare had on the empire, and that the fall of Rome was more of an economic event than a geopolitical event. Had the rulers of Rome done a better job of management, not treated the treasury as their personal bank, and adopted a laissez faire free market approach to the economy, the world might be a totally different place. For one thing, that 900 year depression known as "The Dark Ages" might have been much shorter, or averted totally. The tentacles of the Roman Church might not have been able to reach in and siphon off tax free wealth, productivity, freedom, and ideas in that span of time. Perhaps the ideas of Adam Smith would have been adopted 800 years earlier for the betterment of society as a whole. While all of this is all speculation on my part, one can only imagine how the world would be if the producers and independent thinkers were always allowed to be free of the yoke of excessive government regulation.

## Stefan Jovanovich says:

The Romans had specie and credit; they did not have currency in the sense of Demand Notes. It is an historicism to suggest that there was anything approaching a reserve currency for the world. Merchants did extend each other credit, but there was no central bank, only a Treasury; and there were no exchanges of sovereign currencies between, for example, the Parthians and the Romans. I would appreciate Jeff's providing citations for his assertion that contemporary classical scholars blame capitalism for the decline of the Roman Empire; the ones whom I admire the most - Peter Green , for example - see the matter as being a failure of Rome to accept enterprise as a substitute for war. In their view the "economy" of Rome was based on conquest, not trade; it followed the Napoleonic model. The legions were expected to be able to turn a profit from their wars; when they didn't, things became difficult. No one in the Roman world considered enterprise to be the appropriate occupation for a gentleman. "Trade" was looked down upon as something literally beneath contempt; it was the province of freed slaves, Gentiles and Jews, not "noble" (sic) Romans. Gibbon, who lived in a society that had similar snobberies, found it easy to blame the Christians for the fall of Rome because, in his day, the evangelical Protestants - the Quakers, for example - were a large part of the contemptible merchant class. He could hardly blame the Society of Friends for the fall of Rome, but he could blame their historical antecedents. It also helps to remember that Gibbon was writing for the applause of a public who saw the House of Hanover as a bulwark against the threats of Popish Spain and France and, closer to home, the Catholic parts of Scotland and Ireland.

You wrote "For one thing, that 900 year depression known as "The Dark Ages" might have been much shorter". The so-called Dark Ages were not a 900 year depression.

Firstly, the Dark Ages never really existed. No one in that period described himself as living in the dark ages. The idea of a dark age is an artificial construct of later so-called historians that wanted to glorify their era at the expense of the previous one.

Secondly, over those 900 to 1000 years, a lot of things happened. This is not an homogenous period. There were phases of recession, but there were also phases of strong economic technological and cultural growth, notably the 12th and 13th centuries

Isn't it fair to say that the 12th and 13th century's were really the dawn of the Enlightenment?

Can you really have the light without the dark? Didn't the "rate of progress" of the world slow during the Dark Ages? Couldn't we argue that there were extended periods of time in which man made little or no progress, and in some cases went backwards for extended periods of time?

When one looks at modern depressions, what are we really looking at? During the 1930's the world suffered immensely, but there were a lot of great fortunes made during the depression.

I'd bet that Peter Lynch of Magellan fame didn't see the 1970's as a bad time. He cut his teeth and made his fortune guiding Magellan during a time frame when the major indexes were in toilet.

Some would consider a forest fire an ecological depression. But a fire can actually strengthen the forest and clear the way for new growth……just like we saw in the 1950/60 time frame or the 1980/90 time frame.

One man's depression is another's opportunity. I guess it's all in how you choose to view it.

## Stefan Jovanovich has a more nuanced view:

Jeff's overall point - that Rome and its Mediterranean civilization collapsed and that, with that collapse, came an extraordinary increase in human misery - is unassailable. The Dark Ages may no longer be the politically correct term, but it is still the appropriate one. In measures of public sanitation, life expectancy, and practical individual liberty (the freedom to travel without threat of robbery, the freedom to buy and sell without threat of extortion), it took European cities 1800 years to get back to the place where Rome, Athens, Antioch, and Alexandria were at the time of Augustus [reigned 27 BCE to 14 CE]. The loss of classical literature alone is incalculable.

Scott continues to have a much more optimistic view of modern history than I do. If you include the totalitarian oppressions in Japan, the USSR, German-speaking Europe, Marxist/Fascist Spain, Italy and the consequent devastation of Part II of The World War, the 1930s are a period that has no equal in human history in terms of awfulness. The idea that this was somehow a cleansing devastation that "strengthened the forest" is astounding.

# Population Genetics, from Bruno Ombreux

I had DNA tests performed by the Genographic Project, a five year project using cutting-edge genetic and computational technologies to analyze historical patterns in DNA from participants around the world to better understand our human genetic roots.

There are two possible tests: one based on the Y-chromosome that goes back from father to father; and one based on mitochondrial DNA that goes from mother to mother.

This is not a genealogical study going back a couple of generations. The purpose is to trace the migrations of our ancestors starting about 200 000 years ago and ending about 10 000 years ago.

I don't have the Y-chromosome results yet, but I have the mtDNA ones. From Mom to Mom 2000 generations back. The results are really interesting and pleasantly unexpected.

I belong to the M* haplogroup. This is an East Eurasian haplogroup (genetic clan), and the first to leave Africa 60,000 years ago. My great-grandmother was a Chinese living in Beijing. It is interesting to discover she was not "typical" Chinese, but special in as much as her maternal ancestors were the first to move to Central Asia, remained there for a long time, then probably moved East towards the Pacific coast within the last few thousand years.

# KISS and a Book Review, from Bruno Ombreux

March 18, 2009 | 2 Comments

The Chair advises us to use simple statistics for market analysis. They are good enough for sure. But I wonder every once in a while: why eschew the sophisticated stuff? It must have been developed for a reason, mustn't it?

Or must it?

Little by little, I am starting to see the light in keeping it simple. Today yet again, I got a glimpse of it while reading Unit Roots, Cointegration, and Structural Change by  G. S. Maddala and In-Moo Kim.

This is a book about complicated and modern stuff that is not being used in the tests posted on DailySpec. I am only half through, but I think I can already comment on it. This is an excellent book. It is very clearly written. Usually, when reading this type of book, I am left with the impression that the author is confused. Not here. Maddala and Kim are clear-thinkers who obviously understand their subject matter, to the point that they are able to write about it in an articulate and insightful way.

They understand it so much that they can distance themselves, warning that these tools are mostly ineffective — not to say bull. What I really appreciated upon reading their conclusions about unit root tests (which is just another word for random walk tests), is that they state without any passing thought for political correctness, that the ones everybody uses, like Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP), should not be used, because of dismal power in small samples. This is based on Monte Carlo simulations from other researchers, whose results they provide in the book. These results are ominous and devastating for these oft-run tests. But they don't stop there. They add that they spent two chapters on answering the question: "Which Unit Root Test?". Whereas a better question should have been "Why Unit Root Tests?". To which the answer would frequently be: "They is no reason to perform unit root tests, they are useless for most purposes". I am exaggerating a bit, their statements are mellower, but that's the gist of it. And I like it a lot. A book about unit root tests saying that they are frequently useless (frequently, but not always, to be fair).

In a similar vein, there is a part on panel data unit roots where they mention that a Fisher meta-analysis test from the 1930s, is as good or better than some clever and modern stuff from the 1990s.

So let's Keep It Simple and Stupid!

For 90% of our needs, a grassroots OLS is just as good and more robust than all this rocket-science 20th Century mumbo.

A final word: this book also strikes a perfect balance in maths. They go deeper than usual textbooks, inasmuch as they don't only provide the formulas for the tests, but also the mathematical intuition behind them. But they don't do the full demonstrations, for which they provide an ample bibliography. This is still graduate to upper graduate maths and econometrics so please don't buy this book if you are looking for an introductory text.

The game of Checkers when played scientifically follows the KISS method. Checkers is a game of utility. If you have a win on the board you execute that win in as few moves as possible. I remember a time when a novice (at times I think I am still a novice!) and my opponent had a King in each opposing double corner and I had three Kings and did not know how to consumate the win. With the help of a beginner's Checker book that win on the board is now 'old hat'. Knowledge is power in about anything.

Chair admonishes the Market trader maintain a hand written manuscript. I have a Checker manuscript and record my games for later reference. Any 'tool' that makes you more effective is valuable.

# Sampling Without Replacement, from Victor Niederhoffer

January 11, 2009 | 5 Comments

Sometimes one sees almost every conceivable variations of prices during the year. It's like an evil genius had a bag of tricks and never had to repeat one exactly the same way. I wonder if it's possible to turn this around and assume that there is a finite number of tricks, possible variations that will occur and then predict that the ones not used yet will eventually be used. This becomes particularly relevant for people who look for repetitions of past patterns and in days like this find that there is nothing similar to it in history. Regrettably, that is true almost every day. There should be some creative ways of testing this.

We could look at market entropy, in an information theoretical sense:

Code every possible pattern in bit form, eg 1110011000111

Measure entropy.

See if the market is maximizing it, this would be the "Second Principle of Market Dynamics".

We could also have a Prigogine's Theorem analog i.e the market is forming patterns that will minimize its entropy production.

As an analogous situation, I believe that slot machines, keno games, etc. have their results or numbers selected by random number generator formulas. I always thought that if one is an expert on the existing formulas or was able to generate a random number generator formula based on a series of outcomes then he could beat the game.

I realize that the casino could easily thwart this in keno but it would take additional work on their part for the slots.

In the stock & futures market, I understand there is some pattern recognition software now available. I have no experience with it.

## James Sogi writes:

Maybe sampling something simple like variance over the last couple days might give one a clue. Volatility clusters, and lack of volatility clusters, and variance of volatility within those clusters or length of the clusters, or the survival rates. Again the replacement issue and the assumption of independence clash. The replacement assumes independence, but a cluster model assumes some correlation.

A related (and very important) topic is the number of stable patterns achievable by a set of interconnected nodes. On this topic, a worthwhile read is Stuart Kauffman. Kauffman's work is rather well presented in a chapter of Deep Simplicity: Bringing Order to Chaos and Complexity , John Gribbin, Random House. 2005. ISBN 1-4000-6256-X.

# Poisson Distributions, from James Sogi

December 31, 2008 | 1 Comment

David Stirzaker's Probability and Random Variables, a Beginner's Guide discusses Poisson distributions  which can be used to determine the probability of some rare independent event in a specific period of time. At higher n it approximates the normal distribution. Applied to the market the issue can be stated, during the time that I am long, what is the probability of a crash in the magnitude of 10-10-08 or over 100 points high to low or greater than 7 standard deviation moves. Of 13 such occurrences since 94, 9 were this last October and November. The prior occurrence was in 2000. The clustering reflects that the events are not truly independent negating one of the assumptions of the Poisson model. The clustering seems to be the more important survival factor rather than the probability distribution model. Thus once these outliers appear, the probability of another occurring might be better modeled by a cluster model rather than a distribution curve. It was shown by Rama Cont of the Ecole Polytechnique in 2005 that volatility clusters. We've argued about the effect of these outliers on the distribution curves, but the two regime analysis might be safer. There are other solutions obviously.

Looking at a search for cluster probability models;

2. A cluster-based probability model has been found to perform extremely well at capturing the complex structures in natural textures (e.g., better than Markov random field models).

Having said this, the vacation trade is markedly dampened, and I wonder if the volatility is wearing off as time passes. Some argue for another volatility event in early or mid 2009 but that would not really fit a cluster model of volatility. There seems to be a nice wall of worry to climb and some nice symmetrical drops and gaps in September trade for the elephants to mirror.

Anecdotally, at the store the other day, I heard a lady loudly commenting on how many private jets were lined up on the runway. I'm heading down to the beach after close and will do the jet count for this year. The indicator didn't work too well last year. I'll include my friend's private jet that comes in on the third. Which brings to mind Monty Python… "I'm not dead yet!"

There is one thing with the Poisson distribution. It converges to Gaussian but not uniformly. Tails converge more slowly than the body. One needs quite a few observations in the tails to reach normality, but there are not a lot, else this wouldn't be tails.

This implies we can't even be sure that these events are outliers. Fat tails may not be fat. They may look so only because we don't have enough observations.

Add to this regime changes, volatility clustering etc… The only solution I can think of is not to use too much leverage.

Sharp Sports Betting by Stanford Wong has a good chapter on Poisson distributions for those less mathematically inclined.

# Bottlenecks? from Vince Fulco

April 12, 2008 | 2 Comments

Assuming one's net speeds are up to snuff, where would the IT pros look next for bottlenecks?  Running IB, R, Excel, with a decent sized data import and IE with about 20 tabs open on an AMD 64 bit 2.4Ghz single core with 3 GB running XP_64 and two monitors on a 500Mhz+ video card feels increasingly sluggish.  Would like to move completely to Linux 100% of the time as my quad core with 4GB never has any processor/memory response hiccups but a few apps are just better on XP (unfortunately); e.g. can not make the jump to Open Office.

## Vincent Andres replies:

1/ Any browser with 20 tabs open is something really heavy (this may well be the main bottleneck). And some "tabs" are very heavy.
2/ Depending on what you do with your IB screen(s), it may also be heavy because of the Internet data refresh.
3/ Data import

As for the dual screen, but I would be surprised if it's 100% managed by the video card.

We probably have all more or less similar tasks. My choice has been to split my main tasks onto several computers. So the computer on which I do my main work (where I'm sitting the most): Open Office + R for little studies + Firefox + email + etc. It's separated from other computers used for the other tasks.

Keep in mind that, although the processor speed is always the highlighted number, our jobs go through a long chain (HDD, RAM, etc.), and all pieces are not always accordingly fast. And also that the more opened tasks, the more interrupts to the processor.

I found that the biggest bottleneck on my machine was the ZoneAlarm firewall and antivirus suite.

I got rid of this horrible piece of bloated junk. It has been replaced by a hardware firewall and an antivirus that is light on ressources yet efficient: Avira.

A DSL router acts as an impregnable firewall, is very cheap and doesn't use any computer ressources since it is sitting outside. If one absolutely wants a software firewall, Comodo is much better than ZoneAlarm. It is free and light.

Another thing to get rid of is Windows auto-update. This can slow the machine.

## Jeff Rollert remarks:

I have found putting all communication (e-mail/IM programs, browsing) on a single machine makes crashes far less likely and speeds things quite a bit.  I also use Outlook 2007 (a pig) and Firefox (with so many tabs open I can't count them).

I will also keep Firefox windows open by subject, with tabs like chapters.  Doesn't help speed, but it does help organize thoughts.

# Most Unpleasant Possible Line, from Nigel Davies

March 26, 2008 | 3 Comments

For a chessplayer it's part of the job to consider that the opponent will try to test you with the most unpleasant possible line of play. But I suspect there's a problem here for countists in that the main thrust of the thinking is in what happens in a 'typical' case.

Nevertheless we have to consider what the plan is when the market does exactly what you fear the most. Is there a plan?

Perhaps the most pernicious scenario is the margin call that acts effectively as a stop loss. Let's call them margin stops. If considered during tests, margin stops would render a lot of the systems useless in the first place. Might as well be betting on the ponies.

Isn't it easier to aim for 10% a year, which is achievable with far less risk, and far less work? I mean, at 10% a year for 20 years, one would probably rank high among the top 1% of market participants. And it is incredible what 10% a year can achieve compounded over 20 years if one keep living expenses reasonable in the meantime.

What matters is the end game, isn't it?

# The US Has a Big “Sale” Sign, from Paolo Pezzutti

March 5, 2008 | 4 Comments

When I moved from Italy to the US last year I asked for advice about the opportunity to buy a house during my three-year tour in this beautiful country. Most of the responses were against buying and I am glad that I followed this advice. At the time, the exchange rate between Euro and US dollar was 1.28 vs the current 1.52 (almost a 19% difference). There was a house for sale in the neighborhood for 450K\$. After one year, the house is still for sale, but this time at 380K\$. Moreover, you have to subtract the 19% due to the more favorable exchange rate. For the equivalent of a small two-bedroom apartment in the suburbs of Rome, you can now buy a four-bedroom house here and still have 350K\$ cash. This situation is not related only to the housing market, but the economy in general. The difference in price between goods and assets in the US and Europe during the past year has become impressive. Whenever I happen to fly to Europe I have some relative or friend asking me to buy and bring a new computer, telephone, videogame, golf club, article of clothing, etc. The same is true for the price of cars. But of course importing a car from the US is not so easy! The same applies to the stock market. For Europeans and for others the US has a big sale sign on the country! Sooner or later these imbalances will be resolved and markets will start working in this direction as investors will find opportunities in this new situation.

Same in Hawaii, the Europeans are snapping up land like crazy. What a good deal, they say. I remember the Japanese doing the same 20 years ago. What a good deal, they said. Many of them bailed out in flames from their excesses.

## Stefan Jovanovich remarks:

This is not the first time. One of the least appreciated of President Grant's many virtues was his insistence that the U.S. capital markets become completely open to foreign investment. That was his primary reason for reestablishing the gold standard for the dollar after the Civil War. During the same period J.P. Morgan & Son established its reputation as our country's preeminent investment bank by urging its European customers to exchange their francs and pounds for dollars after the Panic of 1873. When those investments proved to be stunning successes, Morgan's word became literally as good as gold as far as the bankers in Paris and London were concerned. What is truly sad is that, this time, it is the wise visitors like Paolo, not Americans themselves, who see the historic opportunity.

Because of dollar depreciation, visitors have the purchasing power. Even if American see the buying opportunity, they don't have the purchasing power. Also it seems they are in debt, which makes it difficult to add more debt to take advantage of purchasing opportunities.

I see the purchasing opportunities also. I think I'll buy assets in the States in a couple years. It is cheap. And in the long-term, the USA will be better off than Europe.The US has a better demographic pyramid. It has a lower population density. It has the best universities in the world. Taxes are confiscatory but less than in most European countries.

It is cheaper than Europe and has a better and brighter future. This is a buy.And you are right, this is a historic opportunity. I am trying to micro-manage to time the purchase by waiting a couple years, but maybe I am too greedy.

## Stefan Jovanovich replies:

American non-financial corporations certainly have the ability. They have become self-funding, even for capital expansions. They have less dependence on debt markets and banks than they have had in a generation. But their managements seem to be taking their inspiration from Sewell Avery instead of Sam Walton in terms of their confidence about the country's future prospects.

## Bruno Ombreux explains:

The US and Europe have different perceptions of history. In the US, the traumatic experience was the Great Depression. In Europe, it was the Weimar hyperinflation which led to the rise of Hitler which led to the horror of WW2. The purpose of the EU is to avoid another WW2. That was the founding principle of its predecessor the EEC. The purpose of the ECB is to avoid another Weimar. European are ready to take it on the chin and suffer a lot to avoid any repeat of Weimar or WW2. In the 1990s the French experienced two recessions for the sake of Europe. First they absorbed part of the cost of German reunification through imported deflation. Then they cut spending to meet the Maastricht treaty obligations, while due to low growth they should have run an expansionary fiscal policy. They'll do it again. The German will do it too. Everybody will do it. The rest is posturing in the context of negotiations between goverments, as well as trying to influence the ECB. The ECB is not like the Fed. The ECB has only one mandate. It is price stability. It is very precise: CPI right below 2%. The Fed has two mandates, price stability and economic growth. I never understood why, because there is a macroenomic theorem that you need to have as many policy tools as economic targets. If you you want to control inflation for instance, you need only one tool, that is either monetary or fiscal policy. If you you want to control inflation and growth, you need two tools, that is monetary and fiscal policy. That is the case if the Fed and the government are coordinating actions as they seem to be doing. But then it means the Fed is not independent. You can't control inflation and growth and the currency. Something has to give. The job of the ECB is much easier.

## Paolo Pezzutti extends:

The risk in investing in US assets is not related to the value of the assets in US dollars. For example, buying the depressed and daily hit by bad news financials in the long term is something that will work out to be profitable. The financial system is the engine of the US economy. It simply cannot fail and eventually will recover from its excesses. The question mark lies with the exchange rate between euro and US dollar, which could really impact overall performance as it has done in the past years. However, we are at a point of excessive difference in the purchasing power. For example, if you check on Amazon for book prices or on other sites for electronics, such as iPods or Nintendo, you will notice that they sell an item for 100 Euro on one of the Atlantic and for 100\$ in the US, which is quite impressive.

## Kim Zussman replies:

What happened to no-arbitrage theory? A 40 year old student converts his Euros to dollars, buys iPods in the US, and sells them for Euros back home. Same with real estate. Sells his Amsterdam flat, converts to dollars, and buys a beach house in Venice, California. True, ganja is only legal in CA by prescription — but a 50 Euro visit to Dr Pheelgut fixes that.

# Hidden Semi-Markov Models, from Bruno Ombreux

The recent mention of GARCH reminded me of another complicated acronym which could be interesting to specs: HSMM, which stands for Hidden Semi-Markov Models.

The acronym makes it sound complicated, but it is a very simple concept:

1. The market alternates between different regimes, for instance greed and fear

2. Each regime has its own mean and variance, for instance (positive drift, low variance) and (negative drift, high variance)

3. The switch between regimes is probabilistic (with constant probabilities, which is a simplification — it is only a model, after all)

4. Regime duration, aka sojourn time, is some random variable

The fourth point is what differentiates HMM from HSMM, Markov from Semi-Markov. The later is more general in terms of sojourn time distribution.

I feel this type of model is a welcome addition, not alternative, to GARCH in the trader's toolbox:

1. GARCH fits the facts very well but does not really provide any explanation. HSMM fit the facts and provides a physical explanation, for instance alternating moods of greed and fear. Actually, empirical price behavior is incredibly well explained by this simple alternation of different regimes.

2. GARCH focuses on predicting variance. Therefore it is very useful to option traders. HSMM focus on predicting regimes. That is, they focus on the counter's nemesis: ever-changing-cycles. They are useful to all people concerned with the shelf-life of their edges.

What is the probability to be in one cycle? What is the probability to remain in this cycle? Those are questions that HSMM try to answer. Because they are in their infancy, they often provide more questions than answers. It is an open research area. But they can already shed some light on the matter, and it is not always optimistic.

The best fit to empirical decay of squared returns autocorrelations — a long and wordy statement — is provided by negative binomial sojourn times. One interpretation of the negative binomial distribution, is as a not well behaved generalization of the relatively well behaved Poisson distribution. With Poisson, mean = variance. With the negative binomial, mean < variance. If so, then it is not easy at all to forecast regime duration.

However, it is as interesting to know what one can't trade as to know what one can trade. And it is always possible that some more encouraging results could be achieved by looking at other markets, other timeframes, and other phenomena, e.g. backwardation/contango switches in energy markets.

By the way, I am reading Dr McDonnell's excellent book Optimal Portfolio Modeling. As usual, Phil is clear, clever and very right. Everybody should read the book.

# Le Rogue Trader, from Kim Zussman

February 3, 2008 | 4 Comments

Could our French colleagues comment on these aspects of the Affaire Kerviel:

1. Unappreciated low level trader from humble background

2. Big bank known for sophisticated derivative Quant trading by golden boys from elite technical/math universities

3. Gamed the system by hacking compliance and creating false offsetting trades and deceptive emails

4. Industrious - actually skipping holidays in a country which is a holiday

5. Courageously trading OPM through huge swings, at times extremely profitable, in mere hopes of a bigger bonus

6. Bank chose its best trader to unwind into a multi-week market bottom: Waiting 2 weeks could have again been highly profitable

7. He will be lionized in France, where they appreciate when the rich are victimized. (Come to think of it, there will be a place for him on Hillary's cabinet)

## Bruno Ombreux replies:

Regarding Point 2. Actually, they are not sophisticated. The secret to equity derivatives is having a good marketing department and a good sales force. Some of the stuff Jerome Kerviel was hedging, the things called turbos or clicquet options , are mass-marketed equity derivatives designed to part innocent retail investors from their money. With good salesmen, one parts "sophisticated" pension and hedge funds from their money just as easily, except the sale is not done in the mass media but by building relationships with invitations to industry functions and over glasses of champagne.

Serendipitously, I am reading "The complete arbitrage deskbook" from Stephane Reverre, who happened to be head of index arbitrage and quantitative trading in the Tokyo and NY offices of Socgen. Obviously, he is describing activities he learned at Socgen. They are very basic. Nothing sophisticated. They just require a lot of money and good execution. Good salesmen are really helpful here again, for instance in the equity loan market.

# Leather Jacket, from Ken Smith

Bo Keely wrote me once when I made a comment about an expensive leather jacket that I intended to wear to my trip to his little village in the desert.  Keely admonished me, saying upscale dress was a good way to get mugged when traveling.

Of course if you are traveling first class, staying in five-star hotels, using limos to shuttle between airport and hotel, taxi to shuttle from event to event, then the journey is safe.  But I can't travel first class, I stay in motels not hotels, I walk around to avoid taxi charges, I  use public transportation systems.

So I took Keely's advice — put the leather jacket back on the closet hanger.  I have two fine leather jackets — except one is pigskin, made in China. I do not like pig, did not think of what animal the jacket skin might be when I bought it on sale — just thought of the price, was a steal; except it is pig.

The other jacket is tailored, fits like an Eisenhower. Maybe off a Kentucky Derby animal. Bought it at Nordstrom 30 years ago and it's a grand style to wear.  Only thing about it is I wore it to an AA meeting once and a recovering female alcoholic commented "Jesus, we don't need another leather jacket in this group."  I guess she was making a statement about gay dress, seeing the style for gays at that time was to dress in leather.

## Steve Leslie remarks:

In the movie American Gangster there is a scene where Denzel Washington, who plays the gangster Frank Lucas, wears a mink coat with a mink hat to Madison Square Garden for a heavyweight boxing match. Detective Richie Roberts, played by Russell Crowe, takes pictures of people in the front rows around the ring. This helps Crowe in identifying Denzel as the drug kingpin that he is looking for. Up to this point, Denzel had always kept a low profile, thus allowing him to fly beneath the radar of the police. This one gaffe ultimately leads to a subsequent investigation, arrest and conviction.

In addition to its functional role, clothing clearly acts as a signaling mechanism. Another place where it is may be better to dress down is when taking one's car to the garage for repairs; dressing smartly signals a wealthy person who probably doesn't know his manifold sprocket from his flux capacitor.

## Bruno Ombreux extends:

This is a very European attitude. In France, there is a saying: "pour vivre heureux, vivons cachés." That is: to live happily, live stealthly. I had a great aunt which had a lot of money. She dressed so poorly that one day she went to place Vendôme to one of those luxury jewelers with the intention of buying some trinkets. She was denied entrance by the bouncer: "Sorry Madam, we don't think you can afford the merchandise in this place".

In England, really old blue-blooded money consider it a lack of taste to display wealth. They'll go as far as having domestics wear their new clothes so that the clothes acquire quickly the aged patina that makes them wearable to the wealthy.

# Base 20, from Bruno Ombreux

It briefly studied Breton, a Celtic language surviving in Brittany, an area in Western France. This language has a base 20 counting system which was already used by the Gauls, the Celtic tribes that were conquered by Julius Caesar.

French inherited this Celtic substrate. Soixante is 3*20. Soixante-dix is 3*20 + 10. Quatre-vingt is 4*20. Quatre-vingt-dix is 4*20 + 10. There is also the "Hospital des Quinze-Vingt," created by King Saint Louis for 300 blind people. Quinze-vingt is 15*20 = 300.

Of course, base 20 is not the most efficient for counting. The most efficient base for counting is 12. Twelve is a low number with many divisors: 2, 3, 4, 6. Twelve is much more efficient for speed counting than something like 10.

# Putting Real Estate in Perspective, from East Sider

December 20, 2007 | 4 Comments

A friend from the other coast writes:  "There is a complete collapse in demand. We here in California are almost certainly in the midst of another property slump."

Haven't we seen this movie before? California and nearby boom states see moonshots of value and bull-market property geniuses, followed by cyclical shakeouts, despair, lather, rinse, repeat.

In places where property itself was a central business — Miami condoland, the Inland Empire of SoCal — the ugliness can be protracted. Add in the visibility of the problem, and you have the storyline being promoted by The Thundering Herd and other recession-callers.

Let us look at some national figures.  From today's GDP report: Residential Fixed investment dropped 20.5% (annualized Q/Q change), its contribution to the change in GDP was -1.08%. This data shows how much damage has been done by the contraction in housing. A 20% decline in the sector lopped one percent off GDP. But residential real estate is still "only" ~5% of the economy, and it makes sense to keep that in perspective.

## Stefan Jovanovich dissents:

Guys, I am not saying that this is the end of the world; but what used to be a real estate problem has become a banking problem. The Financial Times says that real estate loans are now 40% of bank assets.

I was all-in only a few months ago and took money out of the market only because I wanted to buy a new business. What convinced me that this was not just another slow-down was the spread between our bank's normal jive talk and what they were actually willing to do. We have had the same business in the same location long enough to have seen our "local" (sic) bank morph from Security Pacific to the B of A, and we have seen literally half a dozen people come and go as the branch managers. This is the first time, however, that no amount of collateral was sufficient to get them to say "yes" to a loan even though they are now dealing with a potential borrower who has no debt! They didn't even bother with the usual soft pass — "Have you considered an SBA Loan?" The silver lining that, in this environment, people are going to be able to "trade up" houses is comforting; but it is pure fantasy if the suppliers of actual credit have frozen up - as I think they have. Of course, it could be my bad breath and my charming manners. We shall see.

## Phil McDonnell says:

"News follows price." After a decline bad news will come out to explain it; after a price rise good news stories will come out to explain it.

To some extent the markets are pretty good predictors of events. That partially explains the relationship. However there is a deeper truth in the News Follows Price saw. Simply put the media needs something to write about. They sell fear, they provide information. It is all in pursuit of market share which ultimately leads to advertising revenue.

An author writes a book if and when  he has something to say. The situation is quite different for a media writer. He HAS to say something everyday. It doesn't matter if he actually has something important to say today. His job is to write something anyway. Ideally it will be something so compelling that you, dear reader or viewer, will stop your busy life and check it out.

One of the classic ways to make up a story every day is to look at the market action and try to explain it rationally. After the fact you can always write keen insights like 'The market went up on higher interest rate fears'. The next day might be, 'The market went down on fears of higher interest rates'. Invariably this leads to reinforcement of the meme of the day. In the 90's it was the dot com meme. Recently it has been the declining real estate meme.

But memes evolve. They evolve because the media is under pressure to make the story new. A new twist or angle keeps the meme fresh and compelling. First it was the real estate bubble has burst. Then it was the sub prime mortgage market collapse. The latter evolved into the liquidity crisis. Then the Fed eased. Oops, better not talk about that too much that could be good news. It is better to milk this meme for all it is worth. Then it was the dollar is crashing. But when a meme has been around for a while it gets stale. The media needs to turn to the 'how bad could this get' angle. Perhaps you have heard the old Johnny Carson jokes. Carson could tell a new 'How cold was it' joke every night for decades.

That is why we are seeing stories about recession. More media is piling onto the meme. This could lead to recession, global warming, nuclear winter and a falling sky! Today's article featured the new angle that real estate prices are inaccurate. It is really worse than they are telling us! The article cited one estimate that 1.5% of houses in Denver might be reported as 10% higher in price than they really are. Do the math. That would result in a .15% overvaluation in the Denver market stats - a fraction of 1%.

Quick! Check the sky! Is it still there? I live in the Seattle area so I can't check the sky, as usual. But I trust it is still there.

## Bruno Ombreux agrees:

News coming after price has often very low (0) information-content. Easily/rapidly written/understood.

- Easily written: one must be amazed how newswriters are able to produce/pick descriptive "models" in often less than 24 hours ! (should they have thought of it some hours before, they would be millionaires !). Or maybe those models are just fake ?

- Easily understood, at least for people confusing understanding and memorization.

- It's either tautological "markets up because they didn't range nor gone down",

- Or completely incantatory.

Just some words from the liturgy, put together. No predictive power, even not explanatory power. But it may _look like_ something. That's enough. A majority of people will be happy with it. Thanks for this good stuff for self-deception.

- The most elaborate form may be a linear model: "things will continue". Tautological, incantatory, linear … what are others forms ?

After a market move there is always an open question : why ? Few people have time/skills/tools/data to count/answer. Even few people have time/knowledge to read a true, but a bit long and complex, explanation. (A frank explanation being most of time: "we don't know"). Though, we need to fill the question's volume, to reassure ourself, keep up appearances.

Better a void/fake explanation than no explanation at all, At least, better than an true explanation beyond ourselves.

After a market move there is always an open question : why ?

The post-mortem explanations of market moves show the huge random element combined with human weakness.

Humans don't want to believe that things happen without a (knowable) reason. Ego, insecurity, uncertainty about death after life. So we ascribe explanations irregardless causation.

This is well exemplified by many of the SP500 moves in response to FED rate announcements this year. On Sept 18, they dropped 50 BP and the market jumped 50 points, because "The FED put is still there" (they will counter market declines). OK, but it is also possible the market could have dropped 50 on the same news, because "The FED sees the economy as sliding into recession", and that they cannot stop it.

Then on Dec 11, they dropped 25 BP and the market tanked (biggest drop on a FED day in recent history), "Because traders were looking for a cut of 50 BP". Yes, but it could also have gone up because the FED determined that recession risk was abating and the original crisis overblown.

Some non-human animal experiments are relevant. Recall the pigeons who were fed after pecking a lever: When the feedings came at random intervals, they began to repeat movements and rotations they thought caused the food to appear - not realizing their dance accomplished nothing. Or the rats with electrodes attached to their tails: One group had levers which stopped the painful shocks, the other's levers worked only intermittently. The rats who couldn't control their stress lost weight, shed fur, and became unhealthy, whereas the ones with control remained normal.

The terrible pain and joy generated by markets and other mostly random gambling is more than enough to bring out the animals, as well as herd them to the chapel on Sundays to ask for explanation.

Save my seat!

# Post-Thanksgiving Reflections, from Victor Niederhoffer

November 24, 2007 | 1 Comment

It's good to remember that stocks are valued based on an infinite stream of cash flows. And any balance sheet losses on assets held just affect the book value temporarily and are lost in the fullness of the sweep of payments for risk and innovations and entrepreneurial ability . Same for whether earnings growth is going to be 1 % or 5% next year. The earnings yield versus bond yield is now at close to an all time high. The yield on risky loans has risen and the cost of debt capital has eased. Presumably if the default rate on subprimes is 10% , it is more than compensated by the increase in yield that such loans would now carry . All this comes to a head with the disruptive move at the close on Wed, down 1.5% in 30 minutes. This reminds one of the breaking in of horses featured in such novels as Monte Walsh where the unbroken horse gives a final leap into the corral fence before shuffling off with the owner paying the debt to Monte.

## Phil McDonnell runs some numbers:

Recently the rate on 30 year Treasuries has fallen from about 5.3% to about 4.45%. This is a decline of about 17%. So if the long term earnings are discounted at the long term rate then a very simplistic back of the envelope calculation shows that the value of stocks should rise by something like 17%. However the reality is that stocks have fallen about 8% from when the rate was 5.3%. Together the 17% increase in value plus the 8% should combine for something like a 25% increase in stock values. Some might argue that a more sophisticated model would use the one year rate to discount expected one year out earnings and a two year rate for two year earnings and so on. That is true. But it is worth noting that all the shorter term rates have fallen even more percentage wise than the long term rate.

## Bruno Ombreux extends:

Another exercise is to look at what happens when earnings are changing over time. In the discount formula, the denominator is a power. As a result, early years are heavily weighted and later years much less so.

Let's value the stream of discounting earnings as a perpetuity, because it is easy. It is earnings/interets rate. Let's use 8% which is reasonnable for a risky asset and in line with drift.

Assuming constant \$10 earnings, the stock is worth 10/0.08 = 125.

Now let's assume that earnings are going to take a hit for the next 5 years.

If earnings are 0 for the next five years and then 10 in perpetuity, the company is now worth 125/(1.08)^5 = 85

This a a 85/125 = 32% drop in the value of the company.

The next few years are very important in valuing a company. It is not surprising that stocks drop on the slightest hint that they could experience troubled times ahead, even if in the long term they are profitable.

## George Zachar cautions:

Notional interest rates are only one factor to consider in calculating the appropriate discount. In the current era of (relatively) low and stable rates, perhaps other variables play a increased role.

What tax rate will those future earnings bear? What is the forward trajectory of the regulatory ratchet? Are currency preferences an issue? Finally, should one use real or nominal rates to discount? That would imply the need to forecast inflation too.

While notional rates remain important, the growing/shifting burdens imposed by Washington, and the increased role of international capital pools, means yields are now one discounting factor among many.

## Alston Mabry concurs:

To extend George's argument: What about projections for forex rates? Liquid capital flows across borders, and many investment equations now must contain a forex conversion factor. Must not non-domestic investors evaluate future cash flow discounted by both rates and currency fluctuations?

## Gregory Van Kipnis raises an interesting point:

There has been always been a dichotomy in market valuation between the earnings discount model approach and the book value approach. If we reduce the current discussion to a P/BV versus a P=PV(E,g,i) model for assessing the market outlook, the following additional point may be important to consider. Is there a relationship between BV, on the one hand, and E and g, on the other? If BV (book value) losses were simply a drop in the net value of bricks and mortar there might not be much of a connection to future reductions in E (earnings) and g (growth in earnings). If on the other hand, much of the loss of BV is the destruction of income earning assets (mortgages and their related derivatives) then E and g are proportionately reduced as well. Since such a large proportion of the S&P earnings is related to the financial services industry the current 'neutron bombing' of the housing sector, and the associated loss of financial BVs, it is likely to translate into a more protracted bear market, I fear.

# Duopoly and Performance, from Victor Niederhoffer

November 20, 2007 | 1 Comment

In an interview in New York Magazine, the two leading CEOs of business in Harlem in the 1970s, Frank Lucas and Nicky Barnes, both depicted in American Gangster and Mr. Untouchable, discuss the merits and demerits of duopoly. They both welcomed the other's competition as it provided information efficiencies to the customers. Wikipedia has a nice article on duopoly where they list some examples of major stock market companies involved in duopoly, Moody's and S&P, Pepsi vs Coca-Cola, Airbus vs Boeing, Sotheby's vs Christie's, Sirius vs XM. I would add a favorite of mine Navteq and Teleatlas in navigation systems maps

I am creating a list of duopolies in the stock market based on their stock market value in an industry. A preliminary hypothesis is that the two duopolists perform better than the average company since they face less competition. I wish I had formulated this hypothesis some years ago as I once was a very substantial holder of Navteq and in true duopoly fashion both their outputs are being bought by competitors after bidding wars.

Many market segments in the technology sector are near-monopolies, for example operating systems (Microsoft) and microprocessors (Intel). As Geoffrey Moore has documented, it is very common for a technology market segment to have an "800-pound gorilla" with 40-70% market share, followed by a "chimpanzee" with 15-25% market share and several "monkeys" with less than 10% market share. In enterprise resource planning software, for example, SAP is the 800-pound gorilla, and Oracle is the chimpanzee. In large corporate databases, Oracle is the 800-pound gorilla, and Microsoft is the chimpanzee.

Such quasi-monopolies occur because of the herdlike behavior of corporate information systems managers. These managers are very cautious and want only proven solutions whose bugs have already been discovered by others and corrected. A common statement in decades past was, "Nobody ever got fired for buying IBM." When these managers feel compelled to buy a product in a new technology category, they watch what their peers are doing. Most who notice other managers flocking to a particular vendor select the same vendor.

As this dynamic continues, the entire support structure of the industry, including consultants, solutions aggregators, and independent software vendors, aligns around the 800-pound gorilla's product. As a result, the gorilla has high profit margins. Chimpanzees and monkeys have much lower profit margins because they lack the gorilla's advantages and in some cases have no way to differentiate their products other than lowering prices.

Other 800-pound gorillas include:
- Cisco Systems (network equipment)
- Applied Materials (semiconductor manufacturing equipment)
- EMC (large storage systems)
- Nokia (cell phones)
- eBay (online auctions)
- Qualcomm (wireless chips and licensing technology)

## Bruno Ombreux remarks:

One way to measure competitive pressure or its lack thereof is the Herfindhal index. Stock perfomance vs Herfindhal index has been looked at before. A Google on "Herfindahl index and cross-section stock returns" yields 15000+ pages.

# Vegetable Oil, from Bruno Ombreux

You can buy a car with a diesel engine and use vegetable oil . I heard some people are able to run a car on a 15% diesel / 85% vegetable oil mix. It is great if you live in the countryside. You can grow your own untaxed fuel.

## George Zachar notices:

State makes big fuss over local couple's vegetable oil car fuel

Decatur resident Dave Wetzel may be in hot cooking oil with the Illinois Department of Revenue, who claim he needs to pay \$244 in back taxes for the gallons of vegetable oil he has been running his Volkswagon car on for the past 5 years.

Wetzel uses recycled vegetable oil, which he picks up weekly from an organization that uses it for frying food at its dining facility.

"They told me I am required to have a license and am obligated to pay a motor fuel tax," David Wetzel recalled. "Mr. May also told me the tax would be retroactive."

## Michael Ott explains:

If you do make your own biodiesel from free used vegetable oil, the cost per gallon for materials is around \$.60.  if you pay someone minimum wage to do all the work, the cost is about \$5 per gallon.  Therefore if you enjoy the work and do it as a hobby, it's a good deal.  It's not a money saver.

Some in the biofuels industry are against road taxes on their fuel, but I think they're necessary because infrastructure is already underfunded.  Plus, once we're making a large share of the fuel in the country, it would be noticeable.

# Yet Another Fed Dipsie Doodle, from Phil McDonnell

October 31, 2007 | 1 Comment

Today we were treated to yet another Fed Day dipsie doodle move. What is remarkable is that this event is so predictable on Fed day. It happens almost every time they make an announcement after their meeting.

Instantly, at the moment of the announcement, the market moves dramatically in one direction. This disruptive behavior causes everyone to second-guess his positions. Thoughts of "maybe a drop in interest rates is really bad" crossed many minds. Today's drop of about 12 points in the S&P cash index in just a few minutes was sufficient to free up lots of short term stock.

Just as swiftly the market zoomed upward. All the lost valuation was restored within just a few minutes. The market even added on some ten more points to break into new high territory for the day. "Hooray, it was really good news after all!" Or was it? As this is written it is about an hour after the event and the market is slowly settling back to its level before the announcement came out.

## Bruno Ombreux recalls:

Ten years ago I visited a bank trading room on a Fed day. They probably have been replaced by robots now, but back then they had human traders waiting for the Fed headline to print on the screen, with one finger on the sell hotkey and one on the buy, and they sent orders within one second of the headline's printing.

This contrasted with my then-employer's practice. Currency and fixed-income traders were forbidden to trade around headlines. After a Fed communique, the trading manager and senior traders left the trading room to meet in a quiet office. They analysed the Fed action to reach a consensus decision within 15 minutes, after which they walked back to the trading room, gave instructions to execution traders and then the whole room went rock and roll.

# Changes in Correlation Over Time, from Alston Mabry

There have been comments from analysts recently about changing correlations, for example, this from Morgan Stanley:

US Equity Derivative Strategy
Getting the Best and Worst of Correlation
October 09, 2007
By Peter Polanskyj, Christopher Metli

Correlation between sectors remains high: Correlation among the various sectors of the S&P 500 remains elevated on average, although off the peaks of late August/early September. Among sectors, recent short-term relationships have in some cases differed meaningfully from longer-term relationships.

Correlations among single stocks within specific sectors are a mixed bag: Several sectors have seen correlations among their constituents drop to relatively low levels, including Healthcare, Food/Beverage/Tobacco, Technology, Media, Software, Consumer Services and Food/Staples Retailing. Several sectors have continued to be highly correlated on an absolute and relative basis. Financials are prominent in that landscape.. Full text

Three years ago, Dr. Castaldo commented that changes in correlation may relate to changes in volatility, due to technical reasons and not to changes in the underlying stochastic process. More recently, an article by Harry Kat at City College of London referred to some of the same research around changes in correlation.

In the spirit of "know your tools" (and correlation is an important tool), these three papers seem most often cited:

Pitfalls In Tests For Changes In Correlations

Brian H. Boyer, Michael S. Gibson and Mico Loretan

Mico Loretan and William B English

Kristin Forbes, Roberto Rigobon

Boyer, Gibson, Loretan (BGL) make algebraic and empirical arguments. They create two randomly-generated series, x and y, with a correlation coefficient p : 0<p<1, and show that the correlation coefficient of a subsample of the two series is proportional to the overall p, as the variance of the subsample is proportional to the overall variance of x. That is, as the volatility of x increases, so does correlation between x and y. Here is a graph that is one take on the overall argument. The graph shows how both volatility (measured as sd of x) and correlation vary together over two randomly-generated and positively-correlated series (x and y).

I have been reading generalist books on statistics written by biostatisticians and social scientists. As a rule, they don't like correlation coefficients. Since these coefficients are symmetric/non-causal, they are useless in advancing scientific knowledge.

In finance too, some people don't like correlation coefficients. See for instance these two articles by Embrechts and Alexander [pdf] . They are making points that are in addition to the ones in the links you kindly provided. Some issues are very ivory-towerish: elliptical distributions or joint covariance stationnarity. Others are more down to Earth: extremes creating "ghost effects" in coefficient estimation.

Anyway, the consensus is that correlation coefficients are not a panacea. Actually, it is better not to use them. If one absolutely wants to use them, rank correlation is not as bad as linear correlation. I feel the first article is dismissing rank correlation a bit too fast on the grounds that analytical complexity hinders further mathematical derivations.

What to use instead of correlation? Both articles are promoting their modern alternative pet method for measuring dependencies (copulas and cointegration, respectively). I prefer instead to follow the social scientists' suggestion and build regression models. The nice thing with regression is that assumptions are clear and easy to check. When assumptions are violated, there is a whole slew of more complicated regressions that can be applied.

## Phil McDonnell suggests:

To use regression instead of correlation is misguided. They are the same! After all the square of rho is the same as R^2 from the regression.

## Bruno Ombreux counters:

Yes, but isn't there more information in regression than in correlation? R-squared only gives the proportion of Y explained by X. The regression coefficients together with their standard errors add more information.

In addition, correlation is symmetric: cor(X,Y) = cor (Y,X). X and Y are playing the same role in regard to any possible explanation or causality. Whereas the regression of Y against X is not the same as the regression of X against Y. These are two regressions lines with different slopes. These create a difference between X and Y, there is an explained variable and an explaining variable. Then adding a time dimension one can introduce causality, like Granger causality .

I think that regression contains correlation but it is not the same concept.  Regression is a procedure that examines a number of different statistics, checks residuals, reformulates the equation if necessary.

Another correlation quirk, from Rene Carmona, "Statistical Analysis of Financial Data in S-Plus" Springer-Verlag 2004, pg 99:

"Problem 2.4 This elementary exercise is intended to give an example showing that lack of correlation does not necessarily mean independence!"

Carmona defines X as N(0,1) and shows that Y, a simple function of abs(X) (thus entirely determined by X) with mean 0, variance 1, is uncorrelated with X.

A did a quick R-script to demonstrate; every run will have a slightly different result, but X and Y are always ~0 correlated -

```X<- rnorm(100000)

Y<- (abs(X)-sqrt(2/pi))/(sqrt(1-(2/pi)))

cbind(X,Y)[1:10,]

mean(X); mean(Y)

var(X); var(Y)

cor(X,Y)

Sample run -

> X<- rnorm(100000)

> Y<- (abs(X)-sqrt(2/pi))/(sqrt(1-(2/pi)))

> cbind(X,Y)[1:10,]

X           Y

[1,] -0.7878436 -0.01665691

[2,] -0.4779746 -0.53069754

[3,]  1.3390446  0.89772859

[4,]  0.3362482 -0.76580698

[5,]  1.3081312  0.84644648

[6,]  1.1859110  0.64369580

[7,] -1.6717642  1.44967611

[8,] -0.3082874 -0.81219113

[9,]  0.5582608 -0.39751106

[10,] -0.2235637 -0.95273902

> mean(X); mean(Y)

[1] 0.001646453

[1] -0.00657469

> var(X); var(Y)

[1] 0.9952368

[1] 1.004244

> cor(X,Y)

[1] -0.001873391

>
```

Also, Carmona does a good job of introducing the copula (mentioned in Dr Ombreux's post) as a generalized correlation, and, earlier in the book, nicely motivates kernel density estimation as a generalized histogram, a tool for exploratory data analysis.

At an S-Plus seminar I attended 7ish years ago, Carmona, one of the instructors, spent much time on the copula. Soon afterward, the concept became "famous" via the work of Dr Li  and others.

# Luck Logic and White Lies, reviewed by Victor Niederhoffer

October 19, 2007 | 1 Comment

This brought to mind an American game that was kept alive by a German company after the American shop that created it folded up. A few months ago, it went back under American control, but the Germans kept the flame alive in the interregnum, because the game has lots of fans and good sales in Germany. This game is Battletech.

The game is set in a deeply rich science-fiction universe, with an evolving storyline and a set of characters one gets attached to. It is a board game played on hexed terrain, with miniatures and dice. It is all about manoeuvring to get into favorable positions. But it is different from chess because:

- Luck is important. Dice affect the probabilility of hitting and getting hit, as well as hit locations.

- There is an incredible number of factors to take into account: terrain type (woods, rolling hills, city, etc.), type of forces under control of players (e.g. fast light unit with long range weapons against heavily armored slow unit), enemy skill and psychology.

I feel a game of Battletech is closer to trading than a game of chess or Backgammon!

Because of the absolutely huge number of factors to take into account, one must rely on heuristics and rules-of-thumb, rather than on mechanical probabilistic plays as in Blackjack. In the markets, luck of the draw and the complex interplay of many factors are also prevalent. I feel this is making a case for moving away from rigid rules and even methodology, toward flexible rules-of-thumb. Move the cursor away from the quantitative toward the qualitative. Even though those trading rules-of-thumb are driven by odds as in Battletech, they ought to be vague enough to accomodate the sheer number of influential parameters and to create the conditions for dealing with a wide variety of situations. A example of a good qualitative rule, grounded in numbers yet flexible enough, is Vic and Laurel's principle of buying into panics.

# A Good R Book, from Bruno Ombreux

I am reading Data Analysis and Graphics Using R by John Maindonald and John Braun, and really liking it so far. There are almost no mathematical formulas, but there is a lot of illustrative R code, so it is perfect for those that don't like the hermetism of some stat books, but can write a computer program. It is also written by two authors who are obviously practitioners of statistics, as opposed to theoreticians, which makes it very practical.

## Denis Vako remarks:

The most practical one gets in programming is when one has source code and a debugger; these two make magic in terms of knowledge discovered, which feels better than a good book. The downside — you see what and how, but sometimes cannot figure out why. I also think statistical software help files are a treasure, e.g. those of NCSS.

# Victor Niederhoffer on The Shape of Life

Recent study of the work of Rudolf A. Raff, including his book The Shape of Life, inspired by the supposition of Galton that there are only a small number of forms that are consistent with life based on biological and physical limitations, has led me to consider the specific fixed forms that a species and a market can take. Many of the fixed forms at the basis of the phyla seem to start with a pipe: a mouth, a gut, and an excretory organ. I find that many times the market displays such pipes. Another line of inquiry are the architectural forms that the market displays. Today, the market action in S&P looks like a cathedral. The study of the shape of life raises many fascinating questions as does the architecture of the market. How they be classified and predicted, is a good starting point.

## James Sogi augments:

Both Weyl and Wolfram consider the basic forms of bilateral symmetry as being intrinsic to natural processes in art and nature. (See Wolfram's artificial leaves). Weyl attributes symmetry to even deeper metaphysical processes. The market's basic bilateral process of bid and ask with two opposing forces of buying and selling tends toward the creation of bilaterally symmetrical forms. This lends itself to many predictive applications and the formation of generally negative correlations within lower time frames. The general rule seems to be negative correlation with bouts of correlation breaking out for limited durations. What is not so regular is the durations of said regimes. Study of endings and durations are more robust than study of new beginnings. In other words it is hard to recognize the new regime when it begins, but one can tell when an existing cycle is long in the tooth. On the counting point, Weyl studied the alternating symmetrical patterns prevalent in ancient art friezes. With a typical pattern coming in 3's or other odd or prime numbers, the bilateral symmetry of the market would tend towards an alternating pattern as well. This has predictive application.

Bilateral symmetry is prevalent in nature and the markets (for instance Lobagola, as Vic and Laurel coined it). But it is not the only form of symmetry. Sea urchins display pentagonal symmetry. Could one find higher forms of symmetry in the markets too?

One obvious market is the oil market. There is a fundamental source of of triangular symmetry in the interplay of heating oil, gasoline and crude oil, tradeable in various crack spreads. Going up one further level, oil arb relationships, geographical, time-based and qualitative, are creating a web of multilateral symmetries that are there for the taking.

Changing subjects but keeping with the symmetry theme, I am wondering about the Magic T theory, which is mentioned on pg. 72 of Vic and Laurel's book. Marty Schwartz was a successful S&P trader. He allegedly was a big fan of this so-called theory, though he didn't invent it. The name Magic T is ridiculous, evoking the worst of technical analysis. But it is some kind of Lobagola/mean-reversion theory. There could be something in it. Yet it is not easy to test.

## Russ Sears ponders a related question:

A question I have asked myself, but have never studied in life forms is "why is it that the hierarchical ancestral classification of families of animals done many years ago (Linnaeus, etc.) was proven uncannily correct by modern genetics DNA research?"

The basic classification system was based on the outstanding/noticeable physical differences in life forms. This was well before the complex understanding of the chemistry of life existed.

Obviously, the divergence from normal of the life form filled a niche and created a branch. But why would the visually noticeable difference matter, as much if not more than the hidden chemical differences. Especially when the hidden differences are often the more fundamental or theoretically obvious difference of successful adaptation.

I suspect that once the more fundamental difference occurs, the visually obvious adaptations and physical evolution occur quickly.

A clear case of death to the unfit would be lack of immunity to disease for example. For a converse example the difference between herbivores and carnivores. Fundamentally is a difference in stomach chemistry, not a outward appearance. However, a well known adaption is that herbivores have eyes on their sides to see more of everything, whereas carnivores have eyes in front to see specific targets.

In other words once the subtle difference occurred did the physical difference form rather quickly. Or did large physical obvious differences come first and the subtle difference taking more time follow.

For a speculator, I propose a analogy for carnivore/herbivores eyes. The optimist seeing a vast sea of potential food, must be alert for the sudden unexpected attack. The starving pessimist must focus on the targeted prey. However, both should understand how the other view differs from theirs. The optimist to learn how to shake the predators when he is in their sight. The pessimist, should understand that the optimist has a more rounded view, to see where the opportunity truly is when it appears to come from out of nowhere.

The two key driving forces of evolution are survival and reproduction. Sometimes these are characterized by the phrases:

1. Survival of the fittest
2. Survival of the s-xiest.

In order for an animal to reproduce it must first identify a mate. For most animals the primary identification sense is eyesight. This is not to exclude other senses. Certainly hearing comes into play in the form of mating calls and territorial calls. Smell, touch and even complex courtship behaviors are all used to identify and woo potential mates. So to answer the question why is there such a strong correlation between the outward appearance of a species and its DNA we need only to realize that to reproduce the animals must first recognize each other!

# Generating Trading Ideas, by Bruno Ombreux

October 9, 2007 | 1 Comment

Counting is great, but without trading ideas to test, it will go as far as a Ferrari without gasoline. Let's number "Generating trading ideas" as step zero in the "Counting in 11 steps" manifesto, a variation on Dr. Steenbarger's Opus.

With this in mind, I took up the study of idea manufacturing. One of the shortest books ever published is James Young's "A Technique for Producing Ideas". It is only 48 pages. It was written in the 1940s by an advertiser. Predictably, it is about generating ideas for advertising. But the process is the same in other fields like science or speculation.

Interestingly, the author starts by drawing a distinction between two types of people. Borrowing from Pareto , he presents the "speculators", creative people with ideas, and the "rentiers", unimaginative people thriving in routine. He wrongly translates "rentier" as stockholder when in fact this word means "bondholder". But this is a minor peeve. The translation might not be so wrong: I believe the word stocks was used for bonds in England in the 19th century. This mention of speculators in a book dedicated to idea production is interesting, but not central. The main thesis is that new ideas are a combination of existing ideas. Since existing ideas are needed to produce new ideas, one must have a big store of existing ideas. There are two sources:

- A deep knowledge of the subject under consideration.
- A knowledge of as many other unrelated subjects as possible.

One needs to be both a specialist and a Renaissance man.

Victor is widely recognized as one of the greatest trading idea generators ever, and this site is a reflection of his personality. It deals with a wide variety of subjects, from hunting to BBQ, while displaying a depth of financial knowledge and experience. It took me a while to understand the greatness of this modus operandi. Young's little book helped me pin it down. I am forever grateful to Victor for his teaching by example, showing how by being both an Homo Universalis and an Homo œconomicus, one can create the ideas that are a prerequisite to counting.

This echoes the advice of Alexander Kotov, in Think Like A Grandmaster, that the best way to study openings is to know something about everything and everything about one thing. And I am also very grateful to Victor for his lessons on such matters.

# Computer Bridge, from Bruno Ombreux

Computers are not very good at bridge. It is easy to fool them. If you don't play by the book, they are at a loss to read your cards. Interestingly, this means that they are better against strong players than against weaker players. Their algorithms are noteworthy, however. They are probabilistic in nature, which is closer to the markets than to chess.

# A Statistical Question, from Bruno Ombreux

September 13, 2007 | 2 Comments

I have two time series A and B with 120 monthly observations each. I want to test whether A's yearly changes predict B's yearly changes. But there are only 10 non-overlapping years. What is the least horrible method that would use overlapping 12-months changes? I am thinking of a bootstrap but looking around, I found mention of the Generalized Method of Moments (aka Generalized Estimating Equations) which looks complicated. Do readers have other suggestions?

## Alex Castaldo replies:

The traditional approach used in the literature (by Shiller among others) is to do a rolling (i.e. overlapping) predictive regression and then correct for the overlap by using Newey-West standard errors (rather than the usual standard errors that regression software normally uses).

Victor and Laurel do not like the Newey-West approach, and the literature has been coming around to their point of view. The problem is that Newey-West is correct asymptotically (that is, as the number of data points goes to infinity) but in these problems we do not have a large amount of data (that is why we are resorting to using overlap). Simulation studies show that in small samples the Newey_West method can be biased.

What is the solution? I don't know; it is an open research problem. There is something called the Hodrick (1992) method which is said to be free from small sample bias. (It is different from the Hansen-Hodrick method). Also you might try to read recent papers on the subject, such as Ang and Bekaert "Stock Return Predictability" (2006) and the references therein.

This is what Stata throws up: package lomackinlay from RePEc

TITLE
'LOMACKINLAY': module to perform Lo-MacKinlay variance ratio test

DESCRIPTION/AUTHOR(S)

lomackinlay computes a overlapping variance-ratio test on a
timeseries. The timeseries should be in level form; e.g., to
test that stock returns vary randomly around a constant    mean,
you consider the null hypothesis that the log price series is a
random walk with    drift. The log price series would then be
given in the varlist. If the assumption of homoskedastic
errors in the process generating the differenced series is not
reasonable,  the robust option may be used to calculate a
variance ratio test statistic robust to    arbitrary
heteroskedasticity. This is version 1.0.5, corrected for errors
in logic    identified by Allin Cottrell.

KW: variance ratio test
KW: random walk
KW: heteroskedasticity
KW: time series

Requires: Stata version 9.2

Distribution-Date: 20060804

# Nice Little Programs, from Bruno Ombreux

Lehrstuhl fuer Rechnerorientierte Statistik und Datenanalyse (RoSuDa) offers some nifty R-related and visualisation programs for free.

# The Best Place in the USA, from Bruno Ombreux

I'll give you my foreigner's viewpoint on the best place in the USA. It is always interesting to have a fresh and candid opinion from an outsider. This is one of the reasons why companies hire management consultants.

My credentials as a European expert on the USA are: I was married with a girl from Boston MA, then I had a girlfriend from Santa Barbara CA for eight years, then a girlfriend from New Fairfield CT for three years. And I traveled for business to places like Los Angeles, Houston, Cleveland. The only really famous American town I don't know is New York, which is odd because that's the city Europeans usually visit first.

In my expert opinion, the best place in the USA is Jacksonville FL. My friends in Boston say that my friends in Jacksonville are trash and I am not associating with the best America has to offer, but I don't agree. The people I know there are marvelous. They are simple, honest, welcoming people. They are not intellectuals like the Bostonians, but who needs intellectuals to barbecue and have a keg party? The Saint John's River is a great river and a sight to behold. Climate is great. Never too warm, never too cold. Food is of the utmost quality.

• Plentiful, easy-to-find barbecue
• Hot weather, yes, but frequent sea breezes
• People are sensible and use air conditioning
• Ponte Vedra Beach Resorts, a great spot, with beautiful beaches, great food and great tennis
• Real estate away from the coast isn't too expensive

I love Jacksonville.

# Long/Short Equity, from Bruno Ombreux

Everybody is developing new factors. Look at me, I am small fry, and I have been investing based on factors. I thought I was alone but recent hedge-fund problems have made it clear that the whole world was doing the same thing. In the world of low-frequency finance, everybody has access to the same data, is using similar tools, hence everybody is getting the same results, which are to buy certain stocks with certain risk exposures and sell some others if you fancy being long/short.

This raises the possibility that we were not paid to perform the economic function of taking risk and collecting risk premia, but merely noise investors in a crowd, with ever more money going to funds investing in the same type of stocks, then going to these stocks, increasing factor returns, confirming historical numbers, increasing allocated money, and increasing fund returns in an upward spiral.

Just riding the wave of a Ponzi scheme is not a pleasant perspective. So of course everybody is reassessing factor models.

To be fair, everybody knew he would fail in a liquidity crisis. Quoting the 2004 Wiley book, Pairs Trading, by Ganapathy Vidyamurthy:

A scenario-altering, huge macroeconomic event, for instance, relating to interest rates … typically manifests itself in the form of a liquidity crisis. In these situations, the covariance structure breaks down, leading to the breakdown of the model.

# Book Review from Bruno Ombreux

I just finished reading, How to Lie with Statistics, from Darrell Huff. This is a short book and a very pleasant read with a quaint charm. It was written in the 1950s when everybody was smoking, as attested by numerous illustrations dotting the pages.

It is targeting a mainstream readership, the "average man," as they called him back then, not the scholar at ease with numbers, trained at avoiding bias. It is about spotting the lie whenever someone is presenting statistics to support a sales pitch, be it from Madison Avenue or Capitol Road. This can be extended to Wall Street and whatever thoroughfare is traversing the Cambridge, Berkeley or Stanford thin air academic publication hubs.

This book is best summarized by its last chapter, which provides a checklist for finding lies in statistics:

1. Who says so?
2. How does he know?
3. What's missing?
4. Did anybody change the subject?
5. Does it make sense?

Who says so?

If a toothpaste manufacturer is showing numbers about how great his toothpaste is compared to competition, they may not be as reliable as the same numbers from an independent assessor.

How does he know?

This is about how numbers were collected. Sample bias. The author makes an interesting point that I haven't read very often: a poll is subject to three biases. There is the classical population bias; there is also a bias in the poll questions, which are a sample of possible questions about the subject; and there is a bias in the answers, which are but a sample of the possible opinions of the person answering the poll.

What's missing?

Very often it is the denominator. We keep hearing about how huge the US debt is. It is huge in absolute terms. Divide it by GDP, and it is not as frightening in line with other countries.

Did somebody change the subject?

A statistical non sequitur. The price of rice in Africa has been going up; this is a sure sign of inflation in the USA.

Does it make sense?

I read recently that four percent of French males and one percent of French females were switchers.

On the topic of books, I recently spent a couple of hours in the investing section of a Borders bookstore trying to get a sense of the dominant memes that might be influencing the "public play", as Bacon put it.

Three themes, each of which I saw in multiple books, were:

1. Trend following is the path to profits
2. A disastrous day of reckoning is coming
3. Commodities are better investments than stocks

# Lender of Last Resort, from Bruno Ombreux

One can challenge the usefulness and efficiency of the Fed when it tries to steer the economy, but it is definitely a very useful institution when credit is too scarce. The role of a lender of last resort in nicely explained in the 19th century book, Lombard Street: A Description of the Money Market.

### Jul

#### 31

Henri Cartier-Bresson, arguably the most powerful street photographer in history, used only a Leica 3 with a 50mm f3.5. I enjoy his photo book very much.

Robert Doisneau is great too.

## Janet Murphy writes:

Eugene Atget photographed Paris years before Cartier-Bresson was even born. Both had a keen eye, and their photographs of Paris embody a quiet beauty that remains timeless.

# If You Think Maria is Risky, from Kim Zussman

It's a good thing we don't have French reporters chatting with Ben.

What about Marie Drucker? She is presenting news every evening — but she is dating a former government member, so no chance for us mere mortals.

# Historical Data from Google Finance, from Bruno Ombreux

Smart Quant Discussion

Google Finance doesn't have an official API yet, but here's a way to get historical data from them. The data go back much further than Yahoo (40 years historical), and it's available in 5 minute increments. Broader, cleaner data than OpenTick (which I've found to be a bit flaky). However it's for equities & fixed income only — not much FX & derivative data. Nevertheless, it's better than Yahoo for a free feed of historicals.

# The Problem with Math, and a Book Review, from Bruno Ombreux

There are two issues. One is description of reality. The other is math.

I have just finished the book "Understanding Scientific Reasoning" by Giere, Sickle and Mauldin. It is shedding some light on the first issue. This book has two main parts:

- Theoretical Hypotheses
- Statistical and Causal Hypotheses

Throughout the book, it is made clear that science doesn't pretend to describe any form of absolute reality, out of Plato's cave, but rather to produce approximate models of world phenomena. It is also made obvious that science is not something static, but rather an always-evolving process.

Reading the first part of the book is enough. The second part about statistics is less interesting. The authors somewhat fail to make it clear that statistical hypotheses are a subset of theoretical hypotheses. Perhaps the reason why the authors are devoting so much of their book to trite statistics is that this subject is difficult for their students. Anyway, the first part is great and providing the following summary for any scientific process:

fit/no fit
REAL WORLD      —————————–

—-    MODEL

|                                                                |

|                                                                |
|                                                                |
observation
reasoning/calculation
experimentation                        with experimental setup
|                                                                |
|                                                                |
|                                                                |
V
V
DATA           <———————————–>   PREDICTION
agree/disagree

There is a "real world" from which data is observed. There is a model describing part of the real world. From the model, a prediction about real world data is derived. The prediction must agree with the data in order for the model to be deemed fitting. A statistical model is nothing but a model expressed in statistical terms, e.g., "real world population is normal". Its predictions are what samples should look like if said model is true. The authors are providing a city map analogy. A map is a model of a city, not the city itself. It makes predictions: "Street A is intersecting Street B". These predictions can be checked by walking in the city; that is collecting real world data.

The terms "models" and "hypotheses" are used interchangeably, because they are one and the same thing. A model can be one hypothesis or a collection of hypotheses. "Hypothesis" makes it even clearer that science is not pretending to touch the essence of whatever the "real world" is.

The book is not dealing with the second issue, mathematics. We can ask ourselves where does math fit in?

So far, the description of scientific processes has not involved math. It doesn't need to. Science doesn't need math. Science has but one constraint: to be grounded in logic. Logic can be deployed in math, but also in language, that is rhetoric, or in graphical illustrations, drawn following strict rules.

Compare these 3 versions of Newton's Law of Gravitation:

1. Plain English

"Any two objects exert a gravitational force of attraction on each other. The direction of the force is along the line joining the objects. The magnitude of the force is proportional to the product of the gravitational masses of the objects, and inversely proportional to the square of the distance between them."

2. Graphical (assuming proportions are respected in the drawing)
______________ r _________________

F -F O—————————-o M m

3. Mathematical

F = g * M * m / r^2

One can see immediately the huge advantage of mathematics: Concision. Economy of space. Economy of ink. It is entirely possible to deal with science in plain English without a single formula, but books would be at least 20 times weightier.

Mathematics are a concise lingua franca. But everything written in the language of math can be rewritten in any human language or in the form of graphical descriptions / geometrical figures; think Chinese ideograms or Egyptian hieroglyphs.

The reason why mathematical descriptions do not correspond to the real world, is that they are used in the process of scientific discovery, which itself never applies to the real world, only to the formulation of hypotheses about this world.

# Round Numbers, by Victor Niederhoffer & Alex Castaldo

Today's break through the 1500 level by the S&P index is the first break through of a round number since Nov-17-2006 (1401.2). It is also the seventh consecutive movement upwards through an hundred level without a fall through an hundred level since Nov-01-2002 (900.96).

A chart of the S&P shows a relatively continuous movement up from 100 in 1980 to 1100 in 1998, then a little backing and filling, and then a rise to 1500 (3/22/2000). This was then followed by a precipitous decline back to 800 (7/23/2002).

This raises all sorts of questions about randomness, continuity, tendency for long runs, the drift, and gravitation.

We thought we'd start by looking at a few of the more obvious ones.

SPX Index Daily Data, 100 point box size

 Date Reference Close Dist. frm Ref. Move 1/2/1980 105.76 11/21/1985 100 201.41 101.41 UP 3/23/1987 200 301.16 101.16 UP 12/26/1991 300 404.84 104.84 UP 3/24/1995 400 500.97 100.97 UP 11/17/1995 500 600.07 100.07 UP 10/4/1996 600 701.46 101.46 UP 2/12/1997 700 802.77 102.77 UP 7/2/1997 800 904.03 104.03 UP 2/2/1998 900 1001.27 101.27 UP 3/24/1998 1000 1105.65 105.65 UP 8/31/1998 1100 957.28 -142.72 DOWN 11/2/1998 1000 1111.6 111.6 UP 12/21/1998 1100 1202.84 102.84 UP 3/15/1999 1200 1307.26 107.26 UP 7/9/1999 1300 1403.28 103.28 UP 8/9/1999 1400 1297.8 -102.2 DOWN 11/16/1999 1300 1420.03 120.03 UP 3/22/2000 1400 1500.64 100.64 UP 4/14/2000 1500 1356.56 -143.44 DOWN 7/14/2000 1400 1509.98 109.98 UP 10/10/2000 1500 1387.02 -112.98 DOWN 12/20/2000 1400 1264.74 -135.26 DOWN 3/12/2001 1300 1180.16 -119.84 DOWN 5/21/2001 1200 1200 1312.83 UP 7/6/2001 1300 1190.59 -109.41 DOWN 9/7/2001 1200 1085.78 -114.22 DOWN 9/20/2001 1100 984.54 -115.46 DOWN 10/25/2001 1000 1100.09 100.09 UP 6/21/2002 1100 989.14 -110.86 DOWN 7/18/2002 1000 881.56 -118.44 DOWN 7/23/2002 900 797.7 -102.3 DOWN 7/30/2002 800 902.78 102.78 UP 10/7/2002 900 785.28 -114.72 DOWN 11/1/2002 800 900.96 100.96 UP 6/16/2003 900 1010.74 110.74 UP 12/29/2003 1000 1109.48 109.48 UP 12/14/2004 1100 1203.38 103.38 UP 3/15/2006 1200 1303.02 103.02 UP 11/17/2006 1300 1401.2 101.2 UP 5/3/2007 1400 1501.31 101.31 UP TODAY 1500

We noticed a tendency for UP's to be followed by UP's (and vice versa) so we tested this with a two by two contingency table (previous move listed at the side):

Transition Matrix
=================

 UP DOWN UP 19 7 DOWN 7 6

Fisher's exact test p=0.19

Although the tendency is there, it does not seem to be statistically significant.

## Bruno Ombreux writes:

A classic runs test yields the same p-value in the 2-tailed case. The one-tailed test has obviously half the p-value. That's 0.097, which, as Tukey would say, "is leaning in the right direction". That's not significant but warrants further exploration.

I have a few questions:

- In these cases, is it legitimate to use a one-tailed test? After all, we suspected UP was followed by UP.

- We are testing on the data used to formulate the hypothesis. It is not good practice but in such a long-term study, there is no choice, is there? Not enough data.

- Aren't we wasting our time anyway, UP followed by UP just being an artifact of the positive drift everybody already knows about? What I mean is that the whole exercise is assuredly non-predictive but raises an interesting philosophical question: one-tailed or two-tailed?

Runs Test

Data: S&P Standard Normal = -1.301, p-value = 0.193 alternative hypothesis: two-sided <==> non-random

Data: S&P Standard Normal = -1.301, p-value = 0.097 alternative hypothesis: less <==> trending

keep looking »