11 Dec
2007
11 Dec
'07
4:19 a.m.
Hello, I have implemented support for the beta distribution for boost::random. I was wondering who I should send it to for examination and possible acceptance into boost? The method is based on the paper "Generating Beta Variates with Nonintegral Shape Parameters" ( http://portal.acm.org/citation.cfm?id=359460.359482) by R. C. H. Cheng, which was the best performing fully general purpose method I could find. It uses a reject method whose success rate varies based on the parameters, but whose worst case expected number of samples is 4. If anyone knows of a better method please let me know. -Jeremy Bruestle