[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: comparison of OLS and Fixed-effects regression coefficients

From   Maarten buis <>
Subject   Re: st: comparison of OLS and Fixed-effects regression coefficients
Date   Fri, 1 Feb 2008 12:49:59 +0000 (GMT)

--- Wencke Gwozdz <> wrote:
> I wonder if there is any possibility comparing regression
> coefficients of coss-sectional and panel analysis. The aim is to show
> that both coefficients for a dependent variable are different.
> So far, I estimated an OLS regression and a fixed-effect regression 
> including the same variables, resulting in lower coefficients within
> the fixed-effect model as we did expect. (the data set and the sample
> is absolutely the same.)
> I know that it is possible to test if coefficients are equal by 
> combination of the "suest" and "test" command. But this does not work
> with cross-sectional and panel regressions together.
> Does anyone know if there exists such a test? Is it realizable within
> Stata?

One take on this is is to look at the entire model. Than the answer is
yes, and it is actually in the standard output of -xtreg- with with the
-fe- option: At the bottom of the output you see a line that starts
with "F test that all u_i=0". When all the fixed effects are zero the
fixed effects model collapses to a normal regression, so this test
compares the fixed effects model with the OLS model. 

Hope this helps,

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

Sent from Yahoo! Mail - a smarter inbox

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index