Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Heckman selection and panel data with fixed effects


From   Mark Copelovitch <copelovitch@wisc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Heckman selection and panel data with fixed effects
Date   Tue, 01 May 2007 20:57:11 -0500

Hi,

I'm working with a panel dataset of 47 countries over 20 years -
unbalanced, with a total of 197 observations.  I'm currently controlling
for selection effects in my data by using propensity score matching as a
preprocessing step (using the psmatch2 Stata add-on module), after which
I include the estimated propensity score in an OLS regression with
panel-corrected standard errors and country fixed effects.  However, I
would like to also run a Heckman selection model as a robustness check
for my results.  It's my understanding that fixed effects are not
possible in the first stage probit of the Heckman model, however,
because of the incidental parameters problem.  I have run the selection
model using regional dummies in place of the country effects (in both
stages), and I get similar results to the pcse/matching results.
However, I am wondering if it is possible to include the country effects
only in the second (OLS) stage of the Heckman model, with the regional
effects in the probit first stage.  Any advice on using fixed effects
with the Heckman model, as well as how to implement it within Stata,
would be greatly appreciated.

Thanks,

Mark Copelovitch

________________________________
Assistant Professor
Department of Political Science
University of Wisconsin-Madison
copelovitch@wisc.edu




*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index