st: RE: Areg, absorb vs. xtreg, fe

["Jacobs, David" <jacobs.184@sociology.osu.edu>]

I don't remember the details and my Stata manuals are at home, but I recall that you will be better off with -xtreg- standard errors.

Dave Jacobs

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Katherine Meckel
Sent: Saturday, March 03, 2012 7:46 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: Areg, absorb vs. xtreg, fe

Hello,

I've seen in the archives that the difference between -areg, absorb- and -xtreg, fe- has been visited before, but I didn't see answer to the following:

there are several ways to obtain the OLS estimator in a fixed effects model--does anyone know which models AREG and XTREG use?  Is it least squares dummy variables (LSDV) or a mean-differenced model, e.g.?

I am also wondering, relatedly, if someone can explain why the -areg- help file says the following:
areg is designed for datasets with many groups, but not a number of groups that increases with the sample size.

The reason I ask is because there are different finite sample adjustments for standard errors in fixed effects models depending on whether you are using LSDV or mean-differencing and I don't think XTREG uses any sample adjustment in their standard errors.

Thanks very much for your help, Katherine

*
*   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/

*
*   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/