--- Brian Sartene <[email protected]> wrote:
> My question is the following: Is it legitimate to run a model on the
> pooled data using fixed effects (for time and each state), and then
> to compare the coefficient on the state poverty variable with
> coefficients derived from cross-sections for specific years
> (estimated without state dummies, since their inclusion would induce
> multicollinearity)?
You've run into a known problem with the fixed effects model, and that
is: all variation at the (in your case) state level is absorbed by the
fixed effect, so there is no variation at the state level to be
explained by variables like state poverty level. Your way of forcing
all variation at the state level in the state level poverty and the
individual level error seems to me the wrong way to solve this
question. The way to explain variation at the state level is to use a
random effects model.
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
All New Yahoo! Mail � Tired of unwanted email come-ons? Let our SpamGuard protect you. http://uk.docs.yahoo.com/nowyoucan.html
*
* 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/
