[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: hausman + nocons
Date sent: Fri, 3 Jun 2005 11:02:14 +0200 (MEST)
From: "Cordula Stolberg" <email@example.com>
Subject: st: hausman + nocons
Send reply to: firstname.lastname@example.org
> Dear All,
> I'm trying to do a Hausman test for fixed effects vs random effects
> and both regressions are estimated without a constant. For the fixed
> effects regression, I used "reg" and added the fixed effect manually
Perhaps I am missing something, but I am not sure what you are
looking to do when you estimate a fixed effects regression with no
The slope coefficients in a fixed effects regression are numerically
equivalent to a regression in which all the variables have been de-
meaned, i.e., centred around their respective sample means. In such
a regression, the constant term just drops out.
Or, to put it another way, when you estimate with regress and add the
fixed effects manually, the estimated constant will depend on which
observational unit is the benchmark unit, i.e., which is the omitted
The constant term reported by xtreg,fe is basically an average of all
the fixed effects. Is the nocons constraint that you want to impose
that this average estimated constant = 0?
> and for the random effects regression, I used "xtreg ..., mle nocons".
> I tried to perform the Hausman test by typing
> hausman fixed random, eq(1:2)
> but got the error message "no common coefficients names (eq:coef),
> nothing to test".
> Is that because I have added the fixed effects manually? And if yes,
> how can I then estimate the fixed effects regression without a
> constant ("nocons" doesn't work)?
> * 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/
Prof. Mark E. Schaffer
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS UK
44-131-451-3485 CERT administrator
* For searches and help try: