re:st: Hausman test

[Ricardo Ovaldia <ovaldia@yahoo.com>]

Thank you Kit.

One more issue is the I get recieve the message:

"the rank of the differenced variance matrix (8) does not equal the number of coefficients being tested (9)..."

The two -xtreg- model are the same. Only difference is the -fe- option in the first model. I get estimates for all the covariates in both models so I am not sure why I am getting this message. Any ideas?

Thank you,
Ricardo
 

Ricardo Ovaldia, MS
Statistician 
Oklahoma City, OK


--- On Tue, 1/3/12, Christopher Baum <kit.baum@bc.edu> wrote:

> From: Christopher Baum <kit.baum@bc.edu>
> Subject: re:st: Hausman test
> To: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
> Date: Tuesday, January 3, 2012, 9:32 AM
> <>
> Ricardo said
> 
> I am interested in modeling cholesterol and because of the
> correlation within families I used -xtreg, re- (GLS)
> regression. However when use a Hausman test to compare the
> fitted model with a -xtreg, fe - model, I get p=0.00024. 
> 
> My understanding is that I should not use the random
> effects model, but should be using the fixed effect. What is
> the implication if I ignore the Hausman test and report
> results from the random efects model?
> 
> 
> 
> According to the verdict of the Hausman test, your included
> regressors in the RE model cannot possibly be considered
> independent of the random effects. Thus the coefficients on
> those regressors are biased and inconsistent, and the RE
> estimates should be trashed.
> 
> Kit
> 
> Research Papers in Economics (RePEc)
> http://repec.org
> 
> 
> *
> *   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/