EVIDENCE ON FAT TAILS.

EVIDENCE ON FAT TAILS. Taleb says that when he first met Benoit Mandelbrot, he asked him why a great mathematician would “would take an interest in such a vulgar topic as finance.” Mandelbrot responded “Data, a gold mine of data.” Mandelbrot did an empirical study of how well the normal distribution applies to movements of prices on the stock exchanges. This article in the Economist describes what he found. Looking at the Dow Jones Industrial Average from 1916 to 2003, a normal distribution would predict that it “should have moved by more than 4.5% on six days; it did so on 366. It should have moved by more than 7% only once in every 300,000 years; in the 20th century it did so 48 times.” There are a lot more big price movements than would be expected with a normal distribution. In other words, Mandelbrot’s empirical study showed that it is not always appropriate to use normal distributions when analyzing financial data. (If you didn’t like the graph in the wikipedia article, the Economist article has a graph showing that the normal distribution understates the probabilities of big movements).