Re: st: xtreg and weights

[Comp Avail Bot <compavail@gmail.com>]

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/