Re: st: Poisson Model with endogenous dummy variable

[Partha Deb <partha.deb@hunter.cuny.edu>]

Maria,

I have code for a negative binomial outcome and binary treatment model 
estimated using maximum simulated likelihood.  It should fit your case well. 
It also estimates models for gamma-distributed outcomes.  You can get the 
package, -treatreg2-, by typing

net from http://urban.hunter.cuny.edu/~deb/Stata

from the command line in Stata.

Best.

Partha


Austin Nichols wrote:

> I think Mullahy published a GMM estimator for this specific case, but
> if you are willing to treat the endogenous dummy as continuous, you
> can
>
> ssc inst ivpois
> probit x otherx* z*
> predict phat
> ivpois y otherx*, endog(x) exog(phat)
>
> until you find a better estimator.
>
> On Feb 20, 2008 11:21 AM,  <agostino@unical.it> wrote:
> > dear all,
> > I would like to estimate a poisson model which includes an endogenous
> > dummy variable. To be more precise, this dummy captures a "treatment
> > effect", hence I would like to model the selection process that
> > underlies it by a probit model, and use the fitted probabilities as
> > instruments for the dummy.
> > Does anyone of you know a Stata command (either built in or not) that
> > allows to estimate such a model?
> > thank you very much for any help
> > Maria
> *
> *   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/
--
Partha Deb
Department of Economics
Hunter College
ph:  (212) 772-5435
fax: (212) 772-5398
http://urban.hunter.cuny.edu/~deb/

Emancipate yourselves from mental slavery
None but ourselves can free our minds.
	- Bob Marley

*
*   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/