Gaussian vs. Mandelbrotian: The Great Intellectual Fraud
Sigh. Him again.
Yes, he's unfortunately worth reading, for the drama and stimulation and calling attention to things that are known but not always fully appreciated. But I wish he wasn't worth reading.
His hero Mandelbrot on the other hand? A real scholar with radical ideas, not a provocateur.
The important ideas in his book are well known by thinkers, if not all practitioners, in the fields he criticizes. But, oh, the insults! And so much of what he says is "Such and such models have flaws! Throw them away and do nothing at all until you have perfect models!" That's not terribly useful if you are, say, running an insurance firm. Or doing science.
Anyway, here's the issue of American Statistician duly taking him to task on the technicalities, though unfortunately without the bombast and mud-slinging.
http://pubs.amstat.org/toc/tas/61/3
Also could somebody explain to me why hacker types like to name-check Popper but never Kuhn? Is it because the Star Trek TNG episode with the Binars actually got it right?
Highly, highly recommend this book to anyone who hasn't taken the time yet. Not all that revolutionary for anyone firmly grounded in the real ins and outs of probability, but surprisingly revolutionary for those who know just enough probability to be dangerous, so to speak.
Full version of the chapter here: http://issuu.com/azureo/docs/the_black_swan/1?mode=a_p
Issuu doesn't have a deep-linking feature.
If anyone wants to know what the ten deutschmark bill looks like:
Why is Taleb getting all this mainstream attention? The finance guys have been aware of fail tail distributions all along.
As Eugene Fama points out on his website... http://www.dimensional.com/famafrench/2009/03/qa-confidence-...
"Half of my 1964 Ph.D. thesis is tests of market efficiency, and the other half is a detailed examination of the distribution of stock returns. Mandelbrot is right. The distribution is fat-tailed relative to the normal distribution. In other words, extreme returns occur much more often than would be expected if returns were normal. There was lots of interest in this issue for about ten years. Then academics lost interest. The reason is that most of what we do in terms of portfolio theory and models of risk and expected return works for Mandelbrot's stable distribution class, as well as for the normal distribution (which is in fact a member of the stable class)."
"Take a random sample of any two people from the U.S. population who jointly earn $1 million per annum."
This is where the author loses me. Where is the randomness of the sample, when there are two items which are interdependent?
Later, he criticizes standard deviation as applied to stocks and bonds (decidedly non-random data), finding fault with the bell curve, rather than the misapplication.
Does this chapter make any more sense in the context of the entire book?
Google Books cut off before he got to explaining what the fraud was. Does anyone know?
He seemed to be complaining about assuming distributions were normal without checking, a simple mistakes that is warned against in any introductory statistics class.
