Re: st: negative binomial model

[Maarten buis <maartenbuis@yahoo.co.uk>]

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


--- On Sun, 20/6/10, Mareike wrote:
> I want to analyze count data. Given that for several
> reasons the Poisson analysis showed to be inappropriate,
> my idea was to run a negative binomial model(using the
> command xtnbreg because I have panel data). But what
> happened was that Stata was giving me one iteration after
> the other (what appeared in the result window lokked like
> the following: 
> Iteration 14721:log likelihood = -53.558237  (not
> concave)
> and after 3 hours and ten thousands of iterations I stopped
> the program without having gotten any result.
> Does anybody of you have an idea of what happened? I didn't
> have the problem when running the Poisson model using the
> same dataset.

It means that the model is too complex for your data, so you
need to simplify your model, either move back to -xtpoisson-
or remove some explanatory variables.

Hope this helps,
Maarten



      

*
*   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/