Hi all
I ran a Hausman test to test the null hypothesis that the extra orthogonality conditions imposed by the random-effects estimator with respect to the fixed-effects estimator are valid. However, I can't reach a conclusion because the chi2 of this test turn out to be negative, because "model fitted on these data fails to meet the asymptotic assumptions of the Hausman test". Stata suggest to use the -suest- command.
The command I'm using is the following
. quietly xtreg lny lne lnk, re
. estimates store ran
. quietly reg lny lne lnk s2 s3 s4 s5 s6 s7 s8 s9 s10
. estimates store fix
. suest fix ran, cluster(sector)
. test [fix1=ran1] [fix2=ran2]
But, in the same way as Stata is unable to generate scores for a model using the command -xtreg, fe-, it is unable to generate them for -xtreg, re-. Do you know which approach should I follow to successfully run this test?
Thanks.
Marta.
________________________________
From "Scott Merryman" <[email protected]>
To <[email protected]>
Subject st: Re: SUEST and Xtreg.
Date Thu, 5 Feb 2004 17:12:11 -0600
________________________________
-suest- requires that the estimation commands have a score option, which -xtreg,
fe- does not. However, you could use -regress- with a set of dummy variables.
Scott
----- Original Message -----
From: "Martha Martinez-Martinez" <[email protected]>
To: <[email protected]>
Sent: Thursday, February 05, 2004 10:30 AM
Subject: st: SUEST and Xtreg.
> Is it possible to use suest to test cross-model hypothesis on fixed
> effects models (xtreg, fe)?
Please access the attached hyperlink for an important electronic communications disclaimer: http://www.lse.ac.uk/collections/secretariat/legal/disclaimer.htm
*
* 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/
