st: Fixed-effects estimation with 3000 dummy variables

[Joachim Wagner <wagner@uni-lueneburg.de>]

Dear All:

I have to estimate a model with a very large number of dummy 
variables (interaction terms made from some 260 industry dummy 
variables and 11 year dummy variables) where I want to control for 
establishment level fixed-effects, too. I generated the dummy 
variables, and using pooled data without fixed effects, all this is 
easily done by using AREG and absorbing the 3000 dummy variables. 
Unfortunately, to the best of my knowledge there is no such program 
as "XTAREG". I transformed my data by subtracting the mean values 
within the industry-year cell, and using these data in a simple  OLS 
estimation without a constant gives the same results as the models 
using AREG (as it should).

Question: Is it correct to use XTREG with these de-meaned data, and 
the option FE, to estimate the fixed effects version of the model? Am 
I missing something obvious?

(As an aside, I know that I could use XI:XTREG here, but the do-files 
will run for many days, and the runs have to be done inside the 
Statistical Office for privacy reasons, and the PC is not allowed to 
run over the weekend ...)

Best, Joachim

Prof. Dr. Joachim Wagner
University of Lueneburg
Institute of Economics
PO Box 2440
D-21314 Lueneburg
Germany

Phone: +49-4131-677-2330
Fax: +49-4131-677-2026
Homepage:  www.uni-lueneburg.de/wifo

*
*   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/