Yes, first I did as you say, but then I run the regression again without
p(h) and the again did all that
.estimates store hetero
.local df=e(N_g)-1
.lrtest hetero . , df(`df')
As far as I understand from the manual, -estimates store- would keep the
estimates from the last regression, i.e. without panels(hetero).
So, I did follow Wiggins, but the error is still the same!
Thanks,
Armen
-----Urspr�ngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Richard
Williams
Gesendet: Donnerstag, 31. M�rz 2005 21:43
An: [email protected]; [email protected]
Betreff: Re: st: Test for panel heteroskedasticity
At 09:36 PM 3/31/2005 +0200, Armen Khachatryan wrote:
>Dear statalisters,
>
>I follow exactly the suggestion of Vince Wiggins to FAQ: Testing for
>panel-level heteroskedasticity and autocorrelation (see please
>http://www.stata.com/support/faqs/stat/panel.html ) to test, not to
>sound strange, for heteroskedasticity after running a xtgls, and I type
>
>.estimates store hetero
>.local df=e(N_g)-1
>.lrtest hetero . , df(`df')
>
>
>I get a scaring error message saying
>
>models hetero specified more than once!
No, it appears you have left out the step of estimating the model without
heteroskedasticity. Check the FAQ again. As things now stand, hetero is
being tested against itself.
*
* 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/
--
No virus found in this incoming message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.8.6 - Release Date: 30.03.2005
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.8.6 - Release Date: 30.03.2005
*
* 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/
