Hi Hendrik,
What you could do is estimate a single model with a dummy variable
for year and interactionterms between all variables and this dummy.
The testing wether these interaction terms are zero (with likelihood
ratio test) is testing whether that variable has a different effect
during the different years.
Maarten
--- hwiig <henrik.wiig@e...> wrote:
> I estimate a Heckman model for two different years. The Hausman
> test indicate the models differs between years, which imply I can't
> pool the observations. However, I would also like to compare each
> coefficients separately.
*
* 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/
