Re:st:individual effects and heteroskedasticity

[David Jacobs <jacobs.184@osu.edu>]

If I understand your question correctly, several solutions may be 
possible.  First, if fixed-effects is a plausible estimator, you can use 
areg and specify robust standard errors.  One should be careful to insure 
that there are sufficient groups to estimate the standard errors, however.

Second, although these estimators differ a bit from their random-effects 
counterparts, population averaged models (xtgee in Stata speak) also allow 
the robust option.  And you can ask for the appropriate xtgee model by 
simply adding the letters 'pa' after a coma at the end of your command 
line.  See the manuals and help files for more detailed instructions.

Dave Jacobs


At 05:56 AM 2/13/2003 -0500, you wrote:

> Dear all,
>
> I also have the same question. I notice that not very much often questions 
> about panel data are answered. I would greatly wellcome your help on this 
> topic.
>
> Thanks!
>
>
>
> >I'm a new user of stata and have a doubt. I want to >incorporate 
> individual effects and heteroskedasticity >but xtgls does not allow for 
> this. Would it be a good >idea to xtdata, re and then use regress? (I 
> have >checked between fixed effects, random effects and >poolability and 
> re is the best). I have read that >xtdata is only for looking for the 
> best model, not to >make inference. If this is not a good idea, how 
> could >I estimate a model with individual effects and >heteroskedasticity?
> >
> >I would appreciate your help sincerely.
> >Raul Gonzalez
>
>
>
> Need a new email address that people can remember
> Check out the new EudoraMail at
> http://www.eudoramail.com
> *
> *   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/

*
*   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/