Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Kit Baum <baum@bc.edu> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Re: xtoverid, cluster(X) |
Date | Thu, 18 Mar 2010 08:00:14 -0400 |
On Mar 18, 2010, at 2:33 AM, Marco wrote: > I would like to run xtreg,fe or xtreg,re with cluster option and need > to run xtoverid test to determine if re is appropriate. I guess I need > to include cluster option when I run xtoverid as below indicated. Am I > right? Many thanks > Marco > > > > xtreg, fe cluster(X) > est store fixed > xtreg, re cluster(X) > > > xtoverid, cluster(X) xtoverid will inherit the VCE settings from the prior RE model, so strictly speaking, you don't have to use the cluster() option on xtoverid. You also do not need to estimate the FE model, as it is not used in this artificial regression "generalized Hausman" test. Kit Baum | Boston College Economics & DIW Berlin | http://ideas.repec.org/e/pba1.html An Introduction to Stata Programming | http://www.stata-press.com/books/isp.html An Introduction to Modern Econometrics Using Stata | http://www.stata-press.com/books/imeus.html * * 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/