On torsdag, okt 31, 2002, at 01:20 America/Montreal, Sangho Kim wrote:
Dear Statalister,
I am wondering if it's possible to use xtreg, fe regression for the
fixed-effects model with indicator variables. For the indicator
variables, I am using job classifications (worker1-worker10).
Especially
I am wondering if I can use xtreg, fe regression in which ALL job dummy
variables (without omitting one of them) are used and a constant term
is
also included. Thank you very much for your effort in advance.
I assume the problem is that there are a lot of dummy variables?
A simple way is to
. egen new_group = group(classification_1 classification_2 .... )
and then
. xtreg y x, fe id(new_group)
xtreg imposes that the mean of the categorical new_group dummies are
zero, so you get a constant term. If you care mostly about the
coefficients on x, this is ok.
Erik
--
Erik �. S�rensen, <http://www.geocities.com/erik_oiolf/>.
phd student (economics), Norwegian School of Economics.
currently visiting Queen's University, Kingston Ontario.
*
* 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/
