Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | "Abdalla, Ahmed" <ahmed.abdalla@kcl.ac.uk> |
| To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
| Subject | st: issues with xtoverid |
| Date | Mon, 3 Feb 2014 13:17:25 +0000 |
Hi
I run a robust Hausman test to see which panel data model is more appropriate for my data:
xtreg ceb wtr, re cluster(firmid)
xtoverid
I get the error message:
Error - saved RE estimates are degenerate (sigma_u=0) and equivalent to pooled OLS
However, when I run Hausman such as:
xtreg ceb wtr, fe
estimates store fixed
xtreg ceb wtr, re
estimates store random
hausman fixed random,sigmamore
//since I get Prob>chi2 =0.0000 (i.e.significant p values , <0.05), I use the fixed effect model.
I don't get any error message !
What is wrong with the robust version ?
Thanks
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/