st: Re: Retrieving coefficients using the xtreg command

["Michael Blasnik" <michael.blasnik@verizon.net>]

If you checked help xtreg, you would find that :

predict fixeff, u

should generate the fixed effects in variable fixeff.

You could then:

bysort firmid: keep if _n==1
keep firmid fixeff

and that would create a dataset with just the fixed effects for each firmid

Michael Blasnik
michael.blasnik@verizon.net

----- Original Message ----- 
From: "John de Figueiredo" <jdefig@MIT.EDU>
To: <statalist@hsphsun2.harvard.edu>
Sent: Wednesday, May 04, 2005 4:04 PM
Subject: st: Retrieving coefficients using the xtreg command



> I'm using Stata8.
>
> I want to run a fixed effects (on firmid) regression using the xtreg 
> command.  There are 10000 firms.
> I then want to used the fixed effects coefficients as dependent variables 
> in a second regression (see David Card, Alan Krueger, Journal of Political 
> Economy, 1992, for example).
>
> I run the regression:
>
> xtreg expend sales rd advertise, fe i(firmid)
>
> Is there a way to retrieve the fixed effects coefficients for each of the 
> 10000 firms using the xtreg command?
>
> If not, is there a way to use the xi:regress command and force it to omit 
> a particular reference case (firmid==10) in order to get the coefficients?
>
> Is there a different (easier) way to do this?
>
> Thanks
> John
*
*   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/