Great data graphic out from Reuters yesterday showing the historical 4 statistical moments (mean, standard deviation, skew, and kurtosis) of the major commodity futures markets. Anyone looking at dabbling in any of these markets would be well served to study this data for a bit. The first realization should be that not a single one of these markets is normally distributed, with all having a kurtosis value over 3 (the value for a normally distributed set of data). That roughly means that you should expect a 100 yr flood type move often. Second, do you really want to dabble in Hog futures, for example, which have extreme outlier events 15 times more frequently than a normal distribution? Or maybe that makes it more attractive for you?
The stop the press number for us? Silver had a 9.79 standard deviation move in a single day on September 23rd (a day after having a 5 sigma move). Can this prove once and for all that financial markets are not normally distributed, bringing into question any and all economic theory based on that assumption? To put the nearly 10 sigma (standard deviation is quoted with the Greek Symbol Sigma, thus often quoted as xx Sigma moves) move into perspective, you would need to see an 8 foot 4 inch man walking down the street in order to witness a 10 sigma move. Or how about a high 110 degrees above normal in Chicago in December?
Even in the financial world, where systematic programs are designed to capture such outlier events, a 10 standard deviation move is still extreme, with many having profit targets set at 10 sigma to capture outlier moves.
The data sorted by the number of standard deviation moves on Sep. 23rd is below, and the full data can be seen here: https://graphics.thomsonreuters.com/ce/ALL-VOL.pdf