Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Humaira Asad <humairaasad@hotmail.com> |
| To | STATA HELP <statalist@hsphsun2.harvard.edu> |
| Subject | RE: st: Windmeijer's corrected se in xtivreg2 gmm2s |
| Date | Mon, 27 Jun 2011 16:39:27 +0000 |
Thank you Kit. Your comments are always a source of adding value and quality to what I am working with. Best wishes, Humaira > From: kit.baum@bc.edu > To: statalist@hsphsun2.harvard.edu > Date: Mon, 27 Jun 2011 10:16:21 -0400 > Subject: Re: st: Windmeijer's corrected se in xtivreg2 gmm2s > > <> > On Jun 27, 2011, at 2:33 AM, Humairat wrote: > > > Thank you Kit for your useful input. The model I am estimating using xtivreg2 gmm2s cluster(id) is a static model and not dynamic. In the subsample N is 45, average T is 4.1 and the total number of observations are 185. > > > > If it is a static model, you don't need DPD methods, and the Windmeijer correction is irrelevant. It pertains to the second-stage VCE of DPD estimates, not to GMM estimates of a static model. The cluster-robust VCE should be consistent, and with 45 firms you have a reasonable number of clusters per the Angrist-Pischke-Adams rule of thumb of 42. > > 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/ * * 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/