When conducting system design, you have to decide what type of order your system has to implement. A limit order allows a pre-specified number of shares to transact at a pre-specified price, but it will not cause an immediate and certain execution. On the other hand, the market order is an order to transact a pre-specified number of shares at market price, which will cause an immediate execution, but is subject to price impact.

Limit orders ensure a "discount" when buying a security, although the time-to-fill and the probability to fill are elements which are key to effectively define the effectiveness and profitability of the system. On the other hand, market orders may have an impact on price. Slippage could affect significantly the overall performance of the system, especially when trading a short term system.

Liquidity refers to these different aspects. Based on these differences, liquidity providers or liquidity takers can advantaged in specific market conditions. Assessing the price impact and slippage in case of a market order is easy, especially when trading a limited number of contracts on the e-mini. Much more difficult and challenging is to assess the probability of execution and time-to-fill of a limit order.

I guess that several variables are to be taken into account, such as volatility. How active is the market in terms of transactions/time unit is also to be taken into account. Therefore, traders have to be careful when designing and subsequently testing their systems to accurately quantify the probability of execution of a limit order before jumping to conclusions about the profitability of their system.

Brian Haag comments: 

Limit orders sell a price discovery option; market orders buy the same option. The process of valuing said option is the crux of micro structure and algorithmic execution research. Harris' Trading and Exchanges is a good trailhead.



 It seems candle-charting technique may soon be of little value, if it ever was, due to the advent of electronic markets. I say that as the "seamless" markets we now have are losing a "true opening' as prices close in the pits, then open a few minutes later. Hence, there are no gaps between the close and opens as we once enjoyed.

Since candlesticks operate on the sole basis of the opening and the close, to form the body, the new electronic way of recording price movement totally redefines the spirit as well as the body of candlestick charting.

I guess Nisson, et al, will have to file a habeas corpus with the exchanges or devise some new glitzy charting technique.

David Goodboy writes: 

Many intra-day traders use candlestick charting. In fact it's the default style on most software packages. I believe your idea applies to daily candles only, wherein the open and close of an intra-day candle are based on a chosen timeframe and not the exchanges. 

Larry Williams replies:

I know many traders use candlesticks, intraday, but there are two problems there.

1) What time period is best in electronic markets where for hours nothing happens and then all four alarms go off? The solution is to not use time-based bars. But then what do the candles mean?

2) More important, when I test candlestick patterns most all of them have no predictive value. The sells are better as buys and the supposed best buy, engulfing bullish, is not even as good as a placebo pattern.

Since putting this view forward in an article in Futures Magazine 16 years ago, I still get the same answers in testing these patterns. They may be a great art form, an adjunct for trading to use in consideration of other factors. But my research does not support using them on their own. I am open-minded to all research of these patterns. There is some literature out there purporting they work. Read and test as I have, however, and I think you would disagree. 

Brian J. Haag writes:

What time period is best in electronic markets where for hours nothing happens and then all 4 alarms go off. The solution is to not use time-based bars. But then what do the candles mean?

This has always been true to an extent. Those who only use time-based measures have been at a disadvantage. It's just that now it's getting worse. 



 I believe hedge fund strategies will be new frontiers in the ETF market. As we are seeing ETFs move into more active strategies, we have already seen the beginning of this trend. The quantitative backdrop for evolution can be found in these articles by hedge fund Bridgewater, on selling beta as alfa and levering betas.

My prediction is that the increased accessibility will make it harder to prosper for hedge funds that are currently selling beta as alpha. In effect I see no reason why it couldn't soon be as easy to access some of these strategies as it is to trade the QQQQ today.

Gordon Haave writes:

Yes, but the whole point of the ability to replicate these funds is that you don't need the lockup, or at least not as much of one. One can short volatility without a 1-year lockup.

From Bill Rafter:

The largest portion of hedge fund money is employed in long-short. Long-short is highly liquid and highly scalable, and could easily endure a zero-day lockup. For example, we have a long-only (in theory, less liquid that long-short) large-cap program that has a zero-day lockup. One might ask why. Our answer is "marketing." Investors (particularly pros) are a lot less reluctant to give you money if they can get out on an instant's notice.

Lockups are really only necessary for strategies such as event-driven or distressed assets. The hedge fund industry mostly uses lockups to keep control of its assets. Recall how the recent ('06) Greenwich-based fund went guts-up and tried to manipulate its reports to shareholders to have the latter miss a redemption deadline.

Brian Haag adds:

If the funds are algorithmically managed, they are a short. Fixed systems die. If the funds are actively managed, they are a short. They will not attract the talent that 2/20 type arrangements will, and will thus be the mark at the table.

This whole "you can replicate any hedge fund strategy by adding beta and a few formulas" meme is no different from the "You can beat Wall Street at its own game!" type hucksterism so prevalent in the late 90s. It's just marketing crapola. While the base idea may be sound, that you don't have to get involved in hedge funds to receive average returns, so what? The only possible outperformance in products like these is relative to managers with subpar returns. It's all just another way for the industry to sell average performance.

Managers who do add alpha are very happy about this whole development. It's another source of edge. One needs to look no further than the "Goldman roll" in commodities to see an example.

Charles Sorkin adds:

I have been offered structured notes (intended to be re-offered to our customers) that pay interest based on the Tremont hedge fund indices. Depending on the degree of index participation desired, investors have the option to have total return floored at zero percent (principal guaranteed, like a bank note). Naturally, the secondary market for such a thing is limited, but it's still better than a hedge fund lock-up. Moreover, the issuer is generally an AA-rated large European bank.

Need to get more aggressive? Just buy 'em on margin…

Henrik Andersson adds:

Some of these structured products, which are particularly popular in Europe, are selling with a participation rate of 100% and no Asian etc. This is strange since it seems you get the put for free; but in these cases the cost of the option is most likely taken from the fees of the underlying funds.



 I can't recommend this book highly enough. It is quite possibly my favorite of all time. Today I read Chapter 13, "Metafont, Metamathematics, and Metaphysics: Comments on Donald Knuth's Article 'The Concept of a Meta-Font,'" subtitled "The Mathematization of Categories and Metamathematics."

The chapter makes a case for the absurdity of the notion that creative processes can be mechanized via parameterization. He invokes Godel's Incompleteness Theorem, which states that no non-trivial formal system can demonstrate the truth of all possible assertions, because a set of those assertions is self-referential with respect to the system in question. The canonical example is, "System X is not powerful enough to demonstrate the truth of Sentence S." If the statement is true, it is false, and if it is false, it is true. This contradiction lies at the boundary of all formal systems.

This is the genesis of ever-changing cycles. The market is a system of systems, which continually generates new states, without ever exhausting the set of all possible states. This "essential incompleteness" is the core property of the family of sets called "productive" in mathematics.

(It is tempting to think that all possible states could be examined via a generative process, but there are two problems with this: in a world of zero-bounded asset prices, there is an infinite set of possibilities. Further, even if it were possible to generate a complete set, it would be impractical to attempt to analyze them all, and then the question would become one of screening which states were "relevant." But Godel's Theorem, stated another way, says that you cannot satisfy both of these goals simultaneously. Further, relevance would almost certainly have to be defined a posteriori.)

But this is not to say that pattern recognition and the search for superior risk-adjusted profits is an impossible one. It merely means that minimizing your degrees of freedom (parameter "knobs") will allow you to find relevance more reliably. And those degrees of freedom should be chosen to find the most matches, so that we are nearing the (unreachable) goal of completeness.

But how to choose our parameters? Statistics are of great use here. Analyzing predictive significance of past states is sufficient if the generative process can be (a priori) known, assumed, or believed. An obvious example would be equity markets. If you believe that the circumstances that led to a 1,000,000% price increase in the past century will continue to exist, then your job is to find past states that have predicted higher prices. This is true for any process or market, and is the key underlying factor in trading and investing success: you must have a model of how your market behaves before you can try to take advantage of said behavior via acting on state information. It is worth noting here that the derivatives expert and cohorts' central argument is that the nature of the generative process can not be known, and that therefore statistical moments and properties are just crutches for those who are fooled.

Finding these past states is of course the real pursuit. Hofstadter argues that it is essential to understand that creative processes do not just follow the "letter of the law," but also the "spirit of the law." This suggests a wide world of potential parameters and approaches that are outside the mainstream.

Hofstadter suggests that in generating states it is more effective to consider conceptual roles that must be satisfied, rather than specific, concrete conditions. These roles are quite abstract, and can be partially or wholly fulfilled. Further, these roles are modular and can exist in more than one simultaneous state (or process). The example here is that of serifs (the "feet" on the bottom of fonts), where larger serifs on the letter "a" most likely mean larger serifs on the letters in the font.

It is easy to draw an analogy to volatility in asset markets. If equities suddenly experience a jump in intraday activity, it is at least worth examining if that is predictive of a similar jump in other markets, as it would affect the possible states of those markets.

I could go on (I probably already have), but I think the "meat" of the question I'd like to ask is, "What behaviors can be examined for predictive significance that are not normally parameterized?" It's there that edge lies.



 Chris Cooper wrote: "I am trying to determine what lessons I should learn about my trading during the past week of large moves in the markets."

If the objective is consistent profits (positive returns), how can this be accomplished under all market conditions?

Persisting stable drift, such as the 8-month period that ended last week, requires long exposure/leverage. But as recently demonstrated, this approach has risk that cannot be controlled while maintaining exposure.

Even more ironic is the plight of those who concluded that the 00-03 bear market was "a big one," which would continue as a rational correction of the prior irrationalities. Or the OMWPS (older white males with pony-tails) still holding PALM and EMC from their days as millionares circa 19 and 99.


Maybe the problem should be reframed: Consistent profits are illogical, because no one can anticipate 2/27's, 911s, nor 3/00s. (OK there were many who got one right; some got two, and even a few all 3. Just as magical as 9 heads in a row.)

The only state of risklessness is death. We reach for risk as we follow the path of the delusional Quixote in the moribund hunt for immortality.

Bill Rafter adds:

Depending on how one defines "consistent profits," they may be achievable to the extent that they are more frequent than random would dictate. We would suggest that the focus should be more on reducing the number or severity of periods of negative returns.

Our experience is as follows:

If your investment universe is large-cap stocks, you are going to be vulnerable to overall market declines. Your only escape is to find a tool/indicator that enables you to change your universe during those periods. Some sector rotation will work to the extent that you will be able to claim positive relative returns, but we don't call that "winning," although Wall Street generally does.

If your universe includes mid-cap, small-cap stocks, and foreign equities, you are going to be less vulnerable to overall declines, particularly if such declines occur over an extended period (e.g. 2000-03). But in a "whoosh" such as the market experienced last week, it is more likely that all will fall somewhat together until the panic subsides. If you can predict the whoosh, then good for you. But if you cannot, and your universe is equities, you must find some of those that will behave somewhat independent of the averages.

Brian J. Haag writes:

I've said this in response to various topics, but I will continue to beat the table on it:

The biggest reason so many people perceive increased correlations is because they look at everything in dollars. That's fine; but then they shouldn't be surprised when correlated moves happen. Any time you buy something for dollars, you are essentially selling dollars (or loaning them out, if you want to get all "swappy" about it). So if, after you have made your trade, people decide they like dollars more than your asset, you will lose. And sometimes they decide they want dollars more than just about anything else –and then almost everything goes down together.

It is extremely instructive when looking at a trade (even if you are relatively high-frequency) to price the asset in question in terms of other assets. For example, how much did US equities drop in terms of Euros? Or in terms of gold? It's not orthodox to think of long stocks/short gold as a hedged trade, but sometimes it is. The key is to have the trade on in the right amount and at the right time (of course, that's the key to every trade). But in any case the trade will have add a different kind of diversification, which is probably what you're really after. Call it a "correlation call."


Resources & Links