Fooled By Randomness

[Borodog]

In researching for a post concerning where mathematical economics goes wrong, I came across this interesting review (http://mises.org/story/2056) of Nassim Nicholas Taleb's Fooled by Randomness, an examination of what went wrong in the 1998 hedge fund fiasco. In reality it is a much larger indictment of mainstream economic science.

[ QUOTE ]
What has gone wrong with the development of economics as a science? Answer: There was a bunch of intelligent people who felt compelled to use mathematics just to tell themselves that they were rigorous in their thinking, that theirs was a science. Someone in a great rush decided to introduce mathematical modeling techniques (culprits: Leon Walras, Gerard Debreu, Paul Samuelson) without considering the fact that either the class of mathematics they were using was too restrictive for the class of problems they were dealing with, or that perhaps they should be aware that the precision of the language of mathematics could lead people to believe that they had solutions when in fact they had none…. Indeed the mathematics they dealt with did not work in the real world, possibly because we needed richer classes of processes — and they refused to accept the fact that no mathematics at all was probably better.

[/ QUOTE ]

I haven't had a chance to order the book yet, but it is promising that others are finally waking up to what the Austrian economists have been saying for decades.

[ QUOTE ]
I am now convinced that, perhaps, most of econometrics could be useless — much of what financial statisticians know would not be worth knowing.

[/ QUOTE ]

[ QUOTE ]
This text is a series of logical thought experiments, not an economics term paper; logic does not require empirical verification (again there is what I call the "roundtrip fallacy": it is a mistake to use, as journalists and some economists do, statistics without logic but the reverse does not hold: It is not a mistake to use logic without statistics).

[/ QUOTE ]

[MidGe]

It is a great book and read.

Whilst it may support what Austrian economists have been saying for decades about other economists, I assure you that the book would be equally as trenchant with regards the Austrian economists viewpoints.


This book could not be inetrpreted in anyway as being remotely supportive of AC. LOL

As I said, a good read, one of my top ten non-fiction books by contemporary authors.

By the way, the author has a web site that can be found here (http://www.fooledbyrandomness.com/). Some extracts of his books and other papers can be found there too.

"Our human race is affected with a chronic underestimation of the possibility of the future straying from the course initially envisioned" - Nassim Nicholas Taleb

[Borodog]

[ QUOTE ]
It is a great book and read.

Whilst it may support what Austrian economists have been saying for decades about other economists, I assure you that the book would be equally as trenchant with regards the Austrian economists viewpoints.

[/ QUOTE ]

Such as? Don't take this the wrong way, but have you read enough Austrian economics to know what the "Austrian viewpoint" is? I doubt it. A lot of people think they do, but they have a bizarre charicature in mind rather than anything that resembles actual Austrian theory. For example, it never ceases to amaze me how people accuse Austrians of assuming that people all behave in perfectly rational (as in the game theoretical interpretation) and logical fashion, when actually it is mainstream economics that suffers from this fallacy, and not the Austrians. People seem to unwittingly conflate the perfect rationality and logical argumentation of the Austrian approach to economics with the assumption that the actors in the theory must always behave perfectly rationally and logically.

[ QUOTE ]
This book could not be inetrpreted in anyway as being remotely supportive of AC. LOL

[/ QUOTE ]

Did I say it could? You do realize that anarchocapitalism is not Austrian economics and Austrian economics is not anarchocapitalism, right? Not everything I post is about the state or a proposed lack thereof. /images/graemlins/tongue.gif

[MidGe]

[ QUOTE ]
Did I say it could? You do realize that anarchocapitalism is not Austrian economics and Austrian economics is not anarchocapitalism

[/ QUOTE ]

Point taken... my bad.

[Zygote]

[ QUOTE ]
...it is a mistake to use, as journalists and some economists do, statistics without logic but the reverse does not hold: It is not a mistake to use logic without statistics).

[/ QUOTE ]

true. do you agree that logic AND statistics is far richer than just logic though?

[AvivaSimplex]

I give it 2.5 out of 4 stars. This amazon review pretty much sums it up:

[ QUOTE ]
This book is comprised of the author's musings on the subject of randomness. Taleb is an academic/classicist type who is embarrassed to be a trader. He brags about not watching TV, reading newspapers or engaging in small talk, and refers to himself in the book as an "intellectual snob".

The problem with "intellectual snobs" is that they have an exaggerated opinion of the importance of their thoughts, lapsing into verbose tangents discussing meaningless minutiae. As a consequence, the author spends 200 pages discussing a topic that could be condensed into a few paragraphs. In fact, I will save you the time and effort by providing you with those paragraphs.

Our minds are not wired to think well in the mathematics of probability, and often times ascribe meaning to events that are actually random.

Funds, traders and investors often believe themselves to be experts as a result of a run of good results, when in actuality those runs were the result of randomness. For example, during the 90's many traders and investors who went long in tech stocks thought they were experts (everyone's an expert on the way up), only to find out during the subsequent crash that they were not qualified to detect major market turns, and lacked the skills and discipline to utilize trailing stops (the laws of probability dictate that every once in a rare while there will be a long run of almost uninterrupted price appreciation). The book discusses several similar examples where traders and funds blew up because they didn't properly understand the risks and became married to positions, even after those positions started crashing.

That's pretty much it. The critical point it is that all of the trading dangers discussed by the author can be completely avoided by extensively testing your system to verify that it has a true statistical edge, rather than just being an artifact of random noise, and employing standard trading rules: always use appropriate stops, never average down, don't get married to a position, be flexible enough to reverse your position in a heartbeat, etc.

Now that wasn't so hard, was it, Taleb?

[/ QUOTE ]

[chrisnice]

Hey Boro, I dont mean to hijack your thread but there is a questionrelated to AC which I have been meaning to ask you for a while. Namely, what are the shortcomings or downside of AC? You seem to be a good scientist and the thing that impresses me the most about solid science is the ability of proponents of a particular theory to highlight why it is that they might be wrong. So what are the shortcomings of AC?

[Borodog]

[ QUOTE ]
[ QUOTE ]
...it is a mistake to use, as journalists and some economists do, statistics without logic but the reverse does not hold: It is not a mistake to use logic without statistics).

[/ QUOTE ]

true. do you agree that logic AND statistics is far richer than just logic though?

[/ QUOTE ]

Depending on the application, of course. I have said many times that Austrian economists do not "eschew empiricism on principle" as its detracters will often mischaracterize it to do. Austrians merely recognize the role that empiricism can play in the study of economics and restrict it to that role.

[Borodog]

[ QUOTE ]
Hey Boro, I dont mean to hijack your thread but there is a questionrelated to AC which I have been meaning to ask you for a while. Namely, what are the shortcomings or downside of AC? You seem to be a good scientist and the thing that impresses me the most about solid science is the ability of proponents of a particular theory to highlight why it is that they might be wrong. So what are the shortcomings of AC?

[/ QUOTE ]

Frankly, I don't see any shortcomings or downside. Anarchocapitalism is really just capitalism, and capitalism works every time it's tried.

The closest I can come is saying that I can't predict exactly how something would work, or even if some industries that we currently have would even exist at all. Probably the biggest is the concept of prisons. Prisons are so ridiculously useless that I just don't see how they would exist in the free market (although that is not an argument that they can't or won't; argumentum ad ignorantium). Frankly, I haven't devoted enough thought to it to try to work out logically what would happen in a free society regarding habitual recidivist aggressors. Probably they would find themselves pushed to the frontiers of civilization, perhaps beyond the arctic circle. But as I've said before, that's more the realm of science fiction than it is logical deduction.

[FlFishOn]

This book is on my list already. I may need to prod the library to get it.

[AaronBrown]

I should make a disclaimer that Nassim is a good friend of mine, and I loved his book. Given your feelings about it, you might consider reading his first book, Dynamic Hedging. The covers the technical details of using the insights in a mathematical trading strategy, without the musing, verbose tangents and meaningless minutiae you dislike.

However, for applying the insights beyond trading, I think you missed a few things. You can backtest forever without "completely avoiding" overconfidence. Remember Bertrand Russell's story about the goose who decides the farmer is his friend because he feeds and protects him every day. He gets a positive backtest result daily over his entire life, but he still gets his head chopped off on Christmas Eve.

Also, Taleb's main focus is not on avoiding these errors yourself, you're right that many people have advocated that. He's a guy who makes a fortune taking advantage of them in other people. His usual bet is that things are less likely to change than people expect, but if they do change they will change by a lot more, and in an unexpected direction.

There's also quite a bit of serious philosophical depth to the book, despite it being a reasonably short, easy read. You don't need that for trading risk management, but it's worth knowing anyway.

By the way, you should avoid the new book, Bourgeoisie Virtues, which covers 1% of the ground in four times as many words, and is only the first of four volumes.

[Borodog]

You should ask him if he knows about the work of Mises and the Austrians. I'd be interested in hearing his thoughts.

[AaronBrown]

I will. I have heard him discuss them in general. You can also ask him yourself if you email:

gamma [at] fooledbyrandomness [dotcom]

http://www.fooledbyrandomness.com/

[Borodog]

Thanks, I will. But I want to pick up his book and actually get through it first.

[AaronBrown]

I applaud that thought, both because it's a great book and because you'll get a more useful response that way.

[AvivaSimplex]

So Aaron, I'm curious about Taleb's trading strategy. It's based on the premise that other traders don't recognize the fat tails of price movement distribution. Given that his book and several others have focused on those tails, clearly all the big money is aware of this issue. Shouldn't the options be repriced by now to reflect this knowledge?

Secondly, isn't his strategy vulnerable to its own black swan, namely, a medium- to long-term reduction in securities price volatility?

[AaronBrown]

The basic strategy is to be short volatility at the center, and long volatility in the tails. So you make money if there is no move, or a much larger move than anyone expects. You lose if there is a large move, but within the market expectation.

Of course, this is not a mechanical strategy applied to all markets all the time. Nassim has sophisticated methods for identifying promising situations and levels.

It's true that other people do variants of this approach, but not enough to destroy all the advantage. Moreover, it's not something easy to incorporate into a larger strategy. You need your investors to know that you will show only moderate returns in quiet markets, and can lose money when markets are volatile but not exceptionally so. The big payoff is your fund should deliver home run returns when most hedge funds are crashing. If you mixed this into a larger strategy, investors would see the return drag from the premiums paid (at least on a risk-adjusted basis) without perceiving the insurance benefit.

Therefore, it tends to be a specialized niche. Trust is very important, it's hard to believe a guy who made his first ten million by trend-following is really going to have the discipline to do something like this, however slick his presentation is. You have to be willing to pass up promising opportunities to keep focusing on the main goal of profiting from the crash. Most traders are trained to do exactly the opposite.

The strategy would not do well if volatility range-traded, moving up and down without ever getting really high or really low. You have to believe that's an unlikely state of affairs for any extended period, or at least that you can identify markets or specific trades that are not likely to do it in the immediate future.