Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Madalina Constantin <con_madalina@yahoo.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Test for heteroscedasticity in panel data in STATA |
Date | Wed, 26 Jun 2013 14:40:29 -0700 (PDT) |
Thanks for the link. I did a lrtest and this is what I got: xtgls dep indep , panels(heterosk) igls estimates store hetero xtgls dep indep , igls estimates store homosk local df = e(N_g) - 1 lrtest hetero homosk , df(104) Likelihood-ratio test LR chi2(104)= -6718.22 (Assumption: hetero nested in homosk) Prob > chi2 = 1.0000 Can I conclude from this that the result is not significant, thus there is no problem of heteroscedasticity? For plotting the residuals I only know folowing command: rvfplot, yline(0) which again doesn`t work for panel data. Is there another possibility? --- On Wed, 6/26/13, Gordon Hughes <G.A.Hughes@ed.ac.uk> wrote: > From: Gordon Hughes <G.A.Hughes@ed.ac.uk> > Subject: Re: st: Test for heteroscedasticity in panel data in STATA > To: statalist@hsphsun2.harvard.edu > Date: Wednesday, June 26, 2013, 11:56 PM > You should take a step back and ask > yourself how heteroskedasticity might manifest itself in > your panel. Since there are various sources of > potential heteroskedasticity, you may need to adopt > different model specifications to test different ones. > > The classic form is panel-level heteroskedasticity but with > 6 years for each of 104 companies you have not got enough > observations to test this properly. There is an FAQ at: > > <http://www.stata.com/support/faqs/statistics/panel-level-heteroskedasticity-and-autocorrelation/>http://www.stata.com/support/faqs/statistics/panel-level-heteroskedasticity-and-autocorrelation/ > > > which is extended in a presentation by Gustavo Sanchez of > StataCorp which you will find at: > > <http://cecip.upaep.mx/investigacion/CIIE/assets/docs/doc00026.pdf>http://cecip.upaep.mx/investigacion/CIIE/assets/docs/doc00026.pdf > . > > In general, you would be best advised to plot or otherwise > examine your residuals and think about whether you can > transform or reformulate your models to eliminate any > obvious heteroskedasticity. > > > * > * 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/ > * * 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/