[R] Re: R 0.64.2 at CRAN

[Peter Dalgaard BSA]

Chong Gu <chong at stat.purdue.edu> writes:

> > help.start()
> 
> starts fine, but some clickable links don't work.  For example,
> clicking on \link{dyn.load} brought on by the search engine gets
> bumped, as the page it's looking for was dyn.load.html, whereas typing
> 
> >help(dyn.load) 
> 
> from the command line brings up the desired page dynload.html.  There
> were several instances like this, but I can't remember the others at
> the moment.
> 
> I don't remember running into the problem with 64.1.
> 
> Is this a bug?

Yes, of course. I don't think it's new, though.

> 
> 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))
[1] 18.253
> mean(rpois(1000,rgamma(1000,2,0.1/0.9)))
[1] 0.224
> mean(rpois(1000,rgamma(1000,2,0.9/0.1)))
[1] 18.129

(and the text needs rewriting too, it's a "compound Poisson
distribution", not "heterogeneous Poisson process", right?)


-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._