[Rd] rbinom silently rounds size argument (PR#1377)
Wed, 13 Mar 2002 17:17:40 +0000 (GMT)
Read the subject again, please.
pbinom(2, size=4.1, p=0.5)
makes no sense, but R 1.4.1 gave the result for (2, 4, 0.5) (and S+6 gives
On Wed, 13 Mar 2002, Thomas Lumley wrote:
> On Wed, 13 Mar 2002 email@example.com wrote:
> > Fixed for 1.5.0. Note that pbinom and qbinom did it too.
> But isn't it right for pbinom and qbinom to do it? P(X<=4.5) is a
> perfectly reasonable quantity and is equal to P(X<=4) if X is binomial and
> (with rounding the other way) the same for quantile functions.
> dbinom, OTOH, is a density wrt counting measure and is not defined for
> non-integers (though one could make a case for zero).
Brian D. Ripley, firstname.lastname@example.org
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272860 (secr)
Oxford OX1 3TG, UK Fax: +44 1865 272595
r-devel 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: email@example.com