st: Re: Rsquared within using id & time dummies. xtreg returns collineality error

["Rodrigo Alfaro" <ralfaro@bu.edu>]

I am not sure what is your purpose, but I suggest to do the following:
* fixed with time-dummies
qui tab id, g(dum)
xtreg vary varx dum*, fe

R-2 (within) is in e(r2). areg does not compute R-2 (within) but can
use robust (White) std error correction.

Rodrigo.

----- Original Message ----- 
From: "Jorge Morgenstern" <jmorgenstern@utdt.edu>
To: <statalist@hsphsun2.harvard.edu>
Sent: Thursday, November 24, 2005 11:07 AM
Subject: st: Rsquared within using id & time dummies. xtreg returns 
collineality error



> Hi everyone,
> I have a cross-section time series model, with panel variable for country 
> and monthly data.
> I'm trying to run a regresion with country and year dummies, to control 
> for idiosyncratic issues.
> I can run the model with "xi: areg vary varx i.id*i.time", but when I try 
> to use the "xi: xtreg vary varx i.id*i.time", but it returns an error 
> saying that the regressors are collinear with the panel variable.
> I was able to run the model using xtreg only with i*time dummies, but it 
> returned different coefficients than those from the areg, then it's not 
> performing what I wanted.
> How can I obtain the Rsquared within for my model?
>
> Regards,
> Jorge.
>
> *
> *   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/
*
*   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/