Duncan Murdoch <[EMAIL PROTECTED]> wrote: Your answer was very helpfull! (in fact, the quantile function of a beta (with arbitrary real parameters) is built-in Matlab).
Thank you very much for pointing out the link between F-laws and beta-laws. Best wishes, Davy. > On Fri, 3 Oct 2003 11:45:31 +0200, [EMAIL PROTECTED] (Davy > Paindaveine) wrote: > > >Hello, > > > >In Matlab, the function finv computes the quantile function of a > >F-distribution. Unfortunately, the command finv(p,v1,v2) only allows for > >DFs v1 and v2 that are POSITIVE INTEGERS. Would anyone have a > >modification of this function which allows to deal with fractional DFs? > >(F-distributions are defined in terms of chi-squares, which themselves > >are particular cases of Gamma-distributions allowing non-integer DFs...) > > I'm not a Matlab user, so I don't know for sure, but they'd likely > have an incomplete beta function with general parameters. With the > right normalization, that's the CDF of a beta random variable. > Standard books will give you the simple conversion between F and beta, > based on the fact that if A and B are chi-square with degrees of > freedom df1 and df2, then A/(A+B) is beta, and (A/df1)/(B/df2) is F. > > Duncan Murdoch . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
