Cauchy Principal Component Analysis — mentioned by yummyfajitas

Fri, 13 Mar 2015 10:25:10 GMT

So the CLT says the sum of random variables with rapidly decaying tails will approach a normal distribution. There are similar results showing that the sum of slowly decaying random variables approaches a stable distribution: https://en.wikipedia.org/wiki/Stable_distribution This makes the stable distribution the right answer under some circumstances. For different test statistics (e.g. max drawdown), you've got similar fat tailed distributions, e.g. GEV: https://en.wikipedia.org/wiki/Generalized_extreme_value_dist... As an example of how you might use slowly decaying distributions, consider…

Paper and discussion history. A mention is not an endorsement.

Cauchy Principal Component Analysis · HackerPapers discussions

Cauchy Principal Component Analysis — mentioned by yummyfajitas