I need to estimate a two-way fixed effects model employing unbalanced
panel data. What I need is an implementation of the fixed effect
formulation developed by Wansbeek and Kapteyn (J. of Econometrics,
1989). It is my understanding that STATA does not include such a
procedure. Furthermore, I can't employ STATA's one-way fixed effect
procedure by explicitly employing the full set of dummy variables for
one of the indices; each of the indices contain too many groups.
Do you know if such a STATA program is available? Any information you
can provide would be greatly appreciated.
DETAILS: My data set contains roughly 800,000 records with the number
of individuals, indexed by j, roughly equaling 250,000 and the number of
units (e.g., firms) characterized by a second index, k, being close to
40,000. The data is quite unbalanced in that only a few values of k are
observed for each individual. The number of records per individual also
varies somewhat. In addition to the fixed effects, the model only has a
small number of explanatory variables.
My central goal is to estimate the model and back out estimates of the
fixed effects associated with the second index (firms) as well as the
standard errors for each of those estimates. These will be used in a
second-stage analysis.
Any information about a STATA program or other software would be greatly
appreciated.
Hamp Lankford
Hamilton Lankford
Professor of Economics and Public Policy
University at Albany
State University of New York
*
* 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/