see the following:
Mullahy, J. 1997. "Instrumental-Variable Estimation of Count Data Models: Applications to Models of Cigarette smoking Behavior," Review of Economics and Statistics, 79, 586-593.
Windmeijer, F.A.G. and J.M.C. Silva. 1997. "Endogeneity in Count Data Models: An Application to Demand for Health Care," Journal of Applied Econometrics, 12, 281-94.
In addition, check out Windmeijer's web page, where he has a short paper and GAUSS code posted called ExpEnd that estimates dyanmic FE count models analagous to ABond.
Dann
-----Original Message-----
From: Paul Marschall [mailto:[email protected]]
Sent: Wed 5/21/2003 4:45 AM
To: [email protected]
Cc:
Subject: st: IV-Estimator for count dependent variable
Hi,
can anyone tell me, if there's an appropriate IV-estimator for count dependent variables, specially by using panel data? I've tried using xtivreg and xtabond, but I think, the results are rather biased.
Thanks,
Paul
<<winmail.dat>>
