Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Comp Avail Bot <compavail@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: xtreg and weights |
Date | Tue, 12 Oct 2010 16:27:37 +1100 |
Thanks Stas. On 12/10/2010, at 2:58 PM, Stas Kolenikov wrote: > On Sun, Oct 10, 2010 at 10:12 PM, comp Avail <compavail@gmail.com> wrote: >> 1. I’d like to estimate a standard panel data regression model >> using random effects, but with weights that vary across both time and >> panels. >> xtreg won’t allow that. Is there a way to get around it? >> >> >> 2. I’d like to estimate a standard panel data regression model >> and impose some nonlinear restrictions on the parameters of the type >> b1/b2 = c1/c2. Is there a way to achieve that? Hopefully in a way that >> allows weights to be applied. A solution for either fixed effects or >> random effects or both, would be helpful. > > 1. -gllamm- allows for weights to vary both within and between panels. > Of course you'd want to use -xtreg- to provide the starting values. > > 2. Nonlinear constraints make any model extremely complicated. However > you might be able to rephrase your model with additional variables and > linear constraints among coefficients. > > -- > Stas Kolenikov, also found at http://stas.kolenikov.name > Small print: I use this email account for mailing lists only. > > * > * 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/