RE: st: heteroscedasticity test in random effect model in presence ofautocorrelation

["Dawit Lidia" <ghibertmk@hotmail.com>]

I don't think you find test for heteroscedasticity in random effect model. 
However, if you suspect the problem you can bootstrap the model so that you 
can get standard errors robust to heteroscedasticity. There is a test for 
heteroscedasticity in fixed effect model (xttest3).
Dawit


> From: Abhijit Sarkar <sarkar.abhijit@gmail.com>
> Reply-To: statalist@hsphsun2.harvard.edu
> To: statalist@hsphsun2.harvard.edu
> Subject: st: heteroscedasticity test in random effect model in presence of 
> autocorrelation
> Date: Wed, 19 Jan 2005 14:09:27 +0530
>
> Hi All!
>
> My model is a random effect panel model with significant
> autocorrelation. Iterated GLS with autocorrelation does not produce
> the maximum likehood estimates. In this case how do we check for
> heteroscedasticity in the model.
>
> Can we still go ahead with IGLS-LR test as suggested in
> http://www.stata.com/support/faqs/stat/panel.html
>
> or, there is some other method, possibly better.
>
> Thanx in advance.
>
> --
> Abhijit Sarkar
> UGC Senior Research Fellow (Economics)
> Department of Humanities and Social sciences
> IIT Bombay
> Mumbai, Maharashtra
> India, 400076
>
> Alternate E-Mail ID: abhijit.sarkar@iitb.ac.in
> *
> *   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/
_________________________________________________________________
MSN Messenger http://www.msn.no/computing/messenger - Den korteste veien 
mellom deg og dine venner

*
*   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/