Hi Mikkel > > Thanks for discovering this! I did the implementation, and apparently > assumed notation as the referred source. Sorry for this. > I think there is nothing wrong in the source. Only, you adopted a different definition of the random variable represented by the implemented PascalDistribution: 1. In wikipedia, the r-parameter is the total number of failures, 2. In the CM implementation, the Javadoc states that r is the total number of successes, and the PMF is (correctly) derived accordingly. > > We should also check the variance - this might be suffering from wrong > notation as well. > Yes, I'll do that, too! > > Cheers, Mikkel. > > Now is not a good time to work on o.a.c.m.distribution, since Christian is working on MATH-703. Once this is done, I'll commit a patch to correct this. You will be very welcome to review the changes (except of course if you want to alter your baby yourself ;) ). Sébastien
--------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
