Jun

2

 Statistical Analysis of Extreme Values cited by Chair is useful for risk analysis. I've used survival stats on when to expect the next hi vol event and Vince's binomial analysis for leverage analysis.

Reiss and Thomas use a number of methods and begin analyzing the probability of exceeding a certain value in a given time. For example a daily range of over fifty SP or 100 in a year. They use Pareto distributions as a parametric model among other methods. They state Poisson distributions fit binomial distributions in smaller data sets. The book includes software. I'd be interested to do an analysis of expected high vol events in a year or number of days in a month in an event. I think adding weekly or monthly occurrences of extreme highs would be useful as in the last and current hi vol events.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search