A Question of Extremes

There's a piece in today's FT on extreme value theory which, since I am not a subscriber, I can't read. Extreme value theory is a branch of statistics dealing with the extreme deviations from the mean of probability distributions and is useful for trying to assess risk for highly unusual events, such as 100-year floods, earthquakes and the like. In fact anything with a fat tail distribution, and only a few readings in the tail - like financial market fluctuations, and business cycles. Quick-googling I did discover this research team whose leader has a very acceptable Chuck Berry Johnny B'Good bootleg for listening. This article is straigthforward and informative, while this one is more technical. So now even those of us who are locked-out of the FT premium content won't have to miss out on the fun.