Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Rafal Raciborski <rraciborski@stata.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: clustering and fixed effects poisson |
Date | Mon, 12 Jul 2010 08:07:55 -0500 |
Tatyana Deryugina <tatyanad@mit.edu> said: > I tried using the vce(cluster clustvar) option with xtpoisson in Stata > 11.1 (which I do have), but Stata tells me it isn't available (it's > also not in the documentation). Tatyana should use -vce(robust)- rather than -vce(cluster clustvar)-. -xtpoisson, fe- with robust standard errors implements the formula presented in Wooldridge (1999). The formula is a cluster-robust estimate of the VCE in which the id variable specifies the clusters. The id variable is determined by the current settings of -xtset- which in Tatyana's case is a country variable. -- Rafal rraciborski@stata.com References Wooldridge, J. M. 1999. Distribution-free estimation of some nonlinear panel data models. Journal of Econometrics 90: 77-97. * * 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/