I got a Kindle-v1 Christmas 2007 as a gift. It's sorta great and it kinda sucks.

Great that I can load up just about any PDF on it (either through Amazon's email conversion service or via Mobi-book desktop PDF converter) and thus I can carry around a bunch of journal papers for quick perusal.

Great that if I read a review of a book, I can download the first chapter immediately and determine if I like the book, instead of hoofing it to a book store or ordering it from Amazon and discovering the book is a disappointment.

Kinda sucks because the Kindle (v1 at least) has a very fragile screen. I've cracked four of them, and while Amazon has been very good about replacing them, it is still disturbing that an e-ink screen, which is supposed to be more durable than an LED, seems more fragile than my iPhone's LED.

Another issue is that while one's notes/bookmarks/hi-lites are carried over from Kindle to Kindle for documents one loads up oneself, and for books that are purchased from Amazon, it's not so for periodicals. So when you get a new Kindle you find that while you have recovered that electronic copy of Reason, or The Atlantic, your notes/bookmarks/hi-lites have mysteriously disappeared.

Other problems of switching between Kindle and a replacement Kindle (at least for v1): your browser favorites don't carry over, and old transferred material isn't re-indexed for keyword search; i.e. the first time you purchase a book it's indexed for keyword lookup, but after you switch to a new Kindle anything you move over from the old Kindle is not indexed for keyword look up on the new Kindle. Last but not least, there is no file folders or file tagging. You are limited to sorting your documents by last read date, author, or title. Can't sort or organize your documents by fiction or nonfiction, can't classify documents as to subject area, etc. A folder structure/file tagging system was one of the first things that Kindle v1 users asked for in the Kindle support forums, and it disappointed lots of Kindle fans that it wasn't a feature of Kindle v2.

Nevertheless, with all the complaints, I was very glad when Amazon sent me my new replacement Kindle, as I was going through Kindle withdrawal. Using the Kindle app on an iPhone just isn't the same.

Dr. Corso specializes in modeling energy derivatives in APL, A+, J and k



Bill GatesI'm getting an Intel Core Q6666 and 2Gig RAM. Should I get Vista Ultimate 64bit version? Or just the 32bit Business version? This is for heavy number crunching and trading.

Tony Corso explains:

Unless you have a 64bit application [or you're writing you own code with a 64bit compiler] a 64bit OS won't buy you much of anything. And 64bit Vista has far fewer hardware drivers than the 32bit version. In fact, if you are using purchased software, you might find it runs faster under XP-Pro than under Vista.

As to the processor, there is a relatively cheap 2.6GHZ Quad Core Intel chip [about $275 on NewEgg; the 2.8GHz 'extreme' Intel Quad Core is more than twice as expensive]. Get one of those, and get 4Gig of RAM [make sure your motherboard can see it].

Multiple cores won't do much for you if the software isn't designed to use 'em. At the OS level, Vista automatically assigns my Firefox browser and Excel spreadsheet to different cores, and at the application level Excel2007 'auto-magically' multithreads cell recalculations across multiple cores.

And when considering motherboards, the more recent Intel motherboards have hardware RAID controllers built in [data spread amongst Redundant Array of Inexpensive Drives so if any one drive fails your machine doesn't care and your data are safe].

You might want to consider chaining four drives to that controller so that you won't be gnashing your teeth when you get the inevitable middle of the trading day hard drive hiccup.

Matthew Chlapowski adds:

I was investigating the same problem just a few days ago. To answer your question, 64bit Windows Vista is unlikely to offer you any performance improvement with just 2Gig of RAM. You would have to invest in hardware with at least twice that much RAM, if not four times, to see any improvement. Eventually systems will come with that much memory, but there is little software out there that takes advantage of the added power of such a setup, and few motherboards support it. I would just stick with the Business 32bit version right now, and maybe think about upgrading in a couple of years. That is, unless you are willing to pay top dollar for a system that supports at least 8Gig of RAM , and for the software designed to use it.

Definitely do get an Intel quad-core processor like the Q6600 for number crunching. I've seen benchmarks on those processors and they absolutely smoke anything else available. Benchmarks with Fritz 10 Chess (one of the most intensive number crunching applications on the market) showed Intel quad core processors completing calculations at nearly twice the speed of any other processor you can buy. Don't even think about getting anything else.

Naturally, newer and better things are always in the pipeline, but you have to pull the trigger and buy sometime!



 About a year after the Berlin wall fell, I found myself in Bucharest screwing around with the flying hydraulic knife of a continuous casting steel mill my dad had designed a dozen years before. I ran across an old buddy that had switched jobs and was tasked with finding investment opportunities for a FBC fund run by Mike Holland.

We meet in the lobby of the hotel he was staying at and he asks me if I want to go to Warsaw the next day. So we catch a morning Aeroflot flight to Warsaw. I ask the stewardess for OJ and my buddy asks for tomato juice. She pours out half a glass of orange juice and half a glass of tomato juice and then grabs a bottle of vodka that and poured a shot into both.

My buddy says, ''if the default on the breakfast flight is vodka, maybe eastern Europe isn't the next Singapore or Taiwan, maybe eastern Europe is the next Brazil or Uruguay.''

Nigel Davies adds: 

With the death of Maxim Sorkin (after being discharged from hospital) it's difficult not to notice the number of Russian Grandmasters who've died either during or after medical treatment, the two most notable ones being Leonid Stein and Paul Keres.

It's also difficult not to have noticed the number of domestic Russian plane crashes or the fact that my brief visit to a Russian dentist (whilst residing in Israel) came to a sudden end when I realized he'd been partaking in a few glasses of vodka that lunchtime (I ran like the wind). And this guy had been recommended.

Now does anyone want to speculate why India and China are emerging as powerhouse economies whilst the Russian bear seems to be somewhat confused, and at every level? Well here's a big clue, even according to the Kremlin's figures, 12% of deaths in Russia last year were directly attributable to alcohol. 



I propose One of the biggest problems of trading successfully is the in-built work ethnic, that is instilled in “us” at an early age. This may be something which is regional or religious based, but I believe this certainly works against us when looking to apply discipline to allow for a nice gradual equity curve.

Being from the “west” and school’ed, fed and watered from day one in the pursuit to work hard, thus achieve , reproduce and run a successful family, is I believe not conducive to allow the best characteristics and education to become a trader.

To sit for long periods, and hold the gun, during normal working hours, even though you may of been cleaning the gun and burning the midnight oil looking into research , number crunching and the like, and have done the time, still does not prepare the individual for long periods of holding positions without lifting a finger. These positions also, maybe out of the money, in the money, flat for weeks, before a solid explosion in volatility takes off, where unless risk and money management procedures are solidly adhered to, the human conditioning of work work work, starts to take over, and interferes with correct operating procedures to gain the most out of the position. So in reference to discipline I believe this relevance should be centered on, more so because it is what predominantly disables it.

It could be argued regions, where less importance is placed on routine and normal work practices, could produce interesting results , which of course does nothing to take away from the need for a suitable trading plan and approach to trading.

Tony C. adds:

I have often tried to explain to the young uns’ how I can be wrong most of the time and still make money … and that is a problem with trading.

Typically, you go through 16+ years of schooling, where very early on you quickly come to learn that getting anything less than 9 out of 10 right is a sort of mediocre performance, and getting anything less than 7 out of 10 right is failing … and results in an 80 year old practically blind nun, (who inexplicably possess an arm like Koufax’s), nailing you from across the room with a felt eraser …

Which is to say, you become conditioned.

And then you engage in a profession where getting 6 out of 10 right results in great riches, and even getting 4 out of 10 right results in a more than comfortable living [if you gain 61 cents the 40% of the time you’re correct, and only lose 39 cents the 60% of the time you’re wrong, … and you play often enough.]

So, folks that did reasonably well in their main activity in their formative years are fighting 16+ years of conditioning.

“Geez, I’m right only 4 out of 10 times, I must be stupid to be wrong so often … my success is all luck, I’m a fraud”, etc, etc. If only we could all think like baseball players.



An interesting paper is downloadable here. The gist of the paper is that combining 30 days of historic data with implied volatility gives a better forecast of option prices than simple implied volatility. They demonstrate this by calculating the usual stats (root mean squared error of the forecast, etc.) and by running a 'trading contest'. They describe 5 rules (they call them agents) that, at the simplest, used 30 days of returns to generate a forecast of the stocks distribution, and at its most complex, used Bayes rule to combine 30 days of returns data with implied volatility data to make a forecast of the stocks distribution. They then sell (overnight) ''over priced'' options and buy ''under priced'' options. Naturally, the simple rule makes very little money and the sophisticated Bayes rule makes a ton of money

Some things I would have liked to have known that weren't mentioned:

1. They aren't buying and selling straddles, but individual options (i.e. the simulation isn't delta hedged)… and they do not differentiate between how much of their profit is do to a favorable move in the underlying, and how much of the profit/loss is due to correctly predicting the implied volatility.

2. They do not breakdown how many profitable trades were shorting options, and how many were going long options…it makes a difference to me if I make money while I'm long gamma rather than short gamma

One thing that strikes me as making this hard to try to replicate in other markets is the time structure of volatility problem  It may be okay to take 30 days of OEX/S&P futures (or spot) data and use it when deciding about options that expire quarterly, but I'm leery of trying that with options that expire monthly.  Lets say its Feb 1, and you are looking at options expiring March 1. For stock indexes, the prior January's data will not have any, err, ''structural reason'' to be terribly different from the index future performance data from Feb 1 thru March 1.  Not so with say, oil. During January, the prompt futures contract is February. The January data is using data from Jan 1 (a 30 day forward price against the February contract) through Jan 30 (a one day forward price against the February contract). If you believe (as I do) that things get more volatile as one approaches spot, the January data is a biased representation of what one expects the market to do from Feb 1 to Feb 2  (the overnight trading part of the simulation). The January historic data is a blending of variances of 30 day forwards (Jan 1's vs Jan2's observation) and overnight forward prices (Jan 30's vs. Jan31's observation), but we want to ''overnight trade'' an option on a 28 day forward contract (the March option expiring March 1) — the variances don't ''match up''.

Do any statwise people have an idea as to how I might get around this time structure of volatility problem in trying to reconstruct this simulation using oil prices? (Let's leave the seasonality problems alone for the moment).

N.B. Oil savvy specs will realize I've fudged the expiration cycle/dates a little for the sake of clarity

Dr. Alex Castaldo replies:

Academically it is an interesting paper. They were able to derive extremely messy mathematical expressions for the posterior distribution that are new to the option literature as far as I know. (Bayesian stats sounds good until you try to do the math… often you just hit a brick wall and can't get anywhere).

From a practical point of view I share Prof. Corso's concerns. There are a number of real life issues (like term structure, skew, etc) that are left out.

Also I am bothered that they use (page 17) the "mean of the implied volatilities over the 30 days preceding" and the 30 day realized volatility in their Bayesian estimator. The standard opinion in the industry is that the latest implied volatility is the best estimator and that is one of the estimators they claim they can beat.


Resources & Links