Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Re: pprob
From
Philip Ender <[email protected]>
To
[email protected]
Subject
Re: st: Re: pprob
Date
Sat, 2 Oct 2010 08:23:46 -0700
Stephen Sullivan <[email protected]> wrote:
Hello.
Could someone please provide some Stata advice with respect to "pprob"?
I have a dataset, which does not appear to fit the Poisson distribution.
I then ran glm and nbreg commands as suggested
(http://www.ats.ucla.edu/stat/stata/library/count.htm).
To see how the my variable fit the Poisson and negative binomial
distributions, I downloaded and used the nbvargr command.
Nevertheless, something seems strange as all pprob values for the Poisson are 0.
-------------------------------------
Stephan-
The mean value for your data is 47.92857 so the poisson probabilities
for small counts (0 to 11) is smaller than the number of decimal
places in the output. The -nbvargr- program works best with
relatively small mean counts that are encountered in classroom data.
You can try increasing the n() option to 55 or you can subtract a
constant from each of the observations for graphing purposes. Sorry
that you had problems with the program.
Phil
--
Phil Ender
UCLA Statistical Consulting Group
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/