If you do not mind considering random effects, forget about -xtreg, re-,
and use -xtgee- instead! The latter provides a lots of goodies like
various correlation matrices, or robust standard errors with clustering
over panels, just as Prof. Schaffer desired. It also has the d.f.
corrections through -nmp- and -rgf- options. It does rely on the correct
specification of the mean structure though, so you might want to overload
your model quite a bit with regressors to reduce the bias. -link(id)
fam(gauss)- would give you the RE regression, while -fam(binomial)- along
with -link(logit)-, -link(probit)- or -link(cloglog)- would give you RE
logit, probit, or gompit models. Regression for counts (Poisson and
negative binomial) are yet in the sleeve, and for variables that only take
positive values one might want to specify -fam(gamma)- or -fam(igauss)-
options.
--- Stas Kolenikov
-- Ph.D. student in Statistics at UNC-Chapel Hill
- http://www.komkon.org/~tacik/ -- [email protected]
* This e-mail and all attachments to it are not intended to provide any
* reasonable point of view and was transmitted to you in error. It
* should be immediately deleted by all recepients unless they really
* enjoy communicating with the author :). Other restrictions apply.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
