st: estimated dependent variables

[Katharina Michaelowa <katja.michaelowa@pw.uzh.ch>]

Is there an easy way in Stata to include the relevant errors into a 
regression that uses the estimates from some other regression as 
dependent variables?

In my specific case, I have estimated a fixed-effects regression (in a 
country-year panel setting). In a second stage, I would like to see to 
what extent the country fixed effects are explained by time-invariant 
country level variables, with a particular interest in the R2.
The intention is to show that one can safely use a random effects model 
rather than a fixed effects model in the first place, once these 
variables are taken into account (since they explain a very high share 
of the variation in the fixed effects - at least that's what I believe). 
Of course, to show that, standard errors should be correct and I am not 
sure how to implement this in Stata.

Can you help?
Thank you very much in advance!

Regards,
Katja Michaelowa

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/