Re: st: xtzinb

[David Jacobs <jacobs.184@sociology.osu.edu>]

I think (but am not sure) that there is a panel version of zero inflated in 
Green's econometrics package called Limdep.  If not or if you don't want to 
mess with this less than user friendly package, you can cluster on 
time.  The resulting estimates will not be as efficient as they would be 
with a panel estimator, but they probably will be more robust to 
specification errors.

Dave Jacobs

At 04:09 PM 11/11/2003 +0100, you wrote:

> Dear All,
> I am trying to do a negative binomial regression. My dependent
> variable is a count variable, and for almost 30% of the observation the
> dependent variable is zero. I have a panel dataset, therefore I am
> using xtnbreg. However I couldn't find zero inflated negative binomial
> regression possibility for panel data, what would be more appropriate
> given the large number of zero observations. Does anyone know about
> the existence of "xtzinb"?
> Thanks,
> Miriam
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/