From help-request at octave dot org Wed May 18 06:38:32 2005 Subject: Re: betainc.oct, beta_cdf From: David Bateman To: Samir Sharshar Cc: help at octave dot org Date: Wed, 18 May 2005 13:34:33 +0200 Samir Sharshar wrote: > Hello, > > Thanks for your quick answer... But it doesn't resolve my problem... > > I've used an own function (calctau.m) to test the significativity of > the difference of a distribution of computed values against a beta > distribution. > > Results whose are given by calctau.m on a debian machine using octave > 2.1.57 and a winXP using octave 2.1.36 are identical. But they are > not the same on a mac os 10.4.1 using the binary octave 2.1.69 from > hpc.sourceforge.net. > > Any idea ? > > Here is the relevant code : > And on an x86 with 2.1.69 what happens? I suspect the problem is not in the functions you supplied but in the rand and randn functions from sourceforge.. These were fixed in the CVS on Feb 6, but no new octave-forge release has been made since. I'd suggest recompiling the stuff in octave-forge/FIXES and see if that helps... D. -- David Bateman David dot Bateman at motorola dot com Motorola Labs - Paris +33 1 69 35 48 04 (Ph) Parc Les Algorithmes, Commune de St Aubin +33 1 69 35 77 01 (Fax) 91193 Gif-Sur-Yvette FRANCE The information contained in this communication has been classified as: [x] General Business Information [ ] Motorola Internal Use Only [ ] Motorola Confidential Proprietary ------------------------------------------------------------- Octave is freely available under the terms of the GNU GPL. Octave's home on the web: http://www.octave.org How to fund new projects: http://www.octave.org/funding.html Subscription information: http://www.octave.org/archive.html -------------------------------------------------------------