Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | st: RE: xtivreg2, robust error |
Date | Mon, 2 Jul 2012 15:56:03 +0100 |
Ari, > -----Original Message----- > From: owner-statalist@hsphsun2.harvard.edu > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Ari Dothan > Sent: 02 July 2012 11:47 > To: statalist@hsphsun2.harvard.edu > Subject: st: xtivreg2, robust error > > Hi all, > I use xtivreg2 regressions specifying "fe gmm2s bw(3)". This means "robust to autocorrelation but NOT heteroskedasticity", whereas "fe gmm2s bw(3) robust" means "robust to autocorrelation AND heteroskedasticity". > In some cases I get more efficient results (smaller standard > error) when specifying "fe gmm2s bw(3) robust" than without > the "robust" > requirement. > How could this be explained? You can't safely interpret the differences in this way. The theoretical results for efficiency of the 2-step GMM estimator are asymptotic, i.e., as the sample size goes to infinity. What happens in the real world, with finite samples, can be different. HTH, Mark > Thanking you > > -- > Ari Dothan > * > * 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/ > -- Heriot-Watt University is the Sunday Times Scottish University of the Year 2011-2012 Heriot-Watt University is a Scottish charity registered under charity number SC000278. * * 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/