[R] Re: R 0.64.2 at CRAN
ben at eno.princeton.edu
Tue Jul 6 18:28:29 CEST 1999
On 6 Jul 1999, Peter Dalgaard BSA wrote:
> > Also, can someone double check the negative binomial distribution
> > documentation? I raised the question a while ago but no one seems
> > interested. This concerns the composite Poisson scenario: according
> > my algebra, size=(1-prob)/prob instead of its reciprocal as
> > documented.
> It would seem that you're right:
> > mean(rnbinom(1000,2,0.1))
>  18.253
> > mean(rpois(1000,rgamma(1000,2,0.1/0.9)))
>  0.224
> > mean(rpois(1000,rgamma(1000,2,0.9/0.1)))
>  18.129
> (and the text needs rewriting too, it's a "compound Poisson
> distribution", not "heterogeneous Poisson process", right?)
Both of these corrections are right.
(Just out of curiousity--I know that "compound Poisson process" is
the technically correct term. I meant "heterogeneous" in a non-technical
sense, although I can see that "compound" is better. Does "heterogeneous"
in this context have any exact implied technical meaning?)
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
More information about the R-help