st: gmm: moment condition for random-effect poisson w/ endogenous regressor?

[Conan Li <conanli12@gmail.com>]

Hi Statalist,

I am trying to develop a random-effect poisson model like
E(y(it)|x(it),ηi)=exp(X'(it)β+ηi), where ηi is the individual effect,
and X1 in X' is endogenous and instrumented by lag Z1-Z4(it-1).

The stata11 manual (p595) provides the code for producing the sample
moment condition for FE panel with strictly exogenous variables, but
not for the RE panel. How can modify the code to do that?

Once I have the program for RE poisson, in order to include lag
instruments, do I simply put L.Z1, L.Z2, L.Z3, L.Z4 and variables
excluding X1 in the "instruments()" ? But as discussed by Windmeijer
(2000) on the manual (p599), the second lag (L2) of the endogenous
regressor X1 was used as an instrument. Do I have to do this as well?
The moment condition seems quite different from that of the panel
poisson though.

Thanks.

Conan



--
Qinghua Li (Conan)
State University of New York-College of Environmental Science and Forestry
MS Candidate in Environmental Studies
qli16@syr.edu


-- 
Qinghua Li (Conan)
State University of New York-College of Environmental Science and Forestry
MS Candidate in Environmental Studies
qli16@syr.edu

*
*   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/