Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: panel-level heteroskedasticity


From   Tanveer Shehzad <C.T.Shehzad@rug.nl>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: panel-level heteroskedasticity
Date   Fri, 13 Feb 2009 11:27:56 +0100

Clive Nicholas wrote:
> Nola Agha wrote:
>
>   
>>   I see here:
>>
>> http://www.stata.com/support/faqs/stat/panel.html
>>
>> some advice for testing for panel-level heteroskedasticity.  Can anyone confirm
>> that the null hypothesis for the LR test is "no heteroskedasticity"?
>>     
>
> Yes.
>
>   
>> I actually have a dynamic panel data model.  I used the above LR test without
>> the lagged dv.  Does anyone know of a test for heteroskedasticity for dynamic
>> panel models?
>>     
>
> -findit ivreg2-
> -findit ivhettest-
>
>   
Dear Stata listers:

Actually, I also face a very similar problem.
 I have dynamic panel equation like

  xtabond2 growth(x) laggedvalue(x) laggedgrowth(x), ....

I want to check that dependent variable varies on x value, i.e. hetero.
Now the problem is (i) I donot need xtpcse or xtgls because they give
corrected SE, not a hetro test (ii) xtcsd says xtreg results not found
(iii) ivhettest says last estimation results not found (iv) I cannot use
LM test bcoz N>>T (as suggested in xtcsd stata article)

Is there some other solution? Your answer will be really helpfull.

Best Regards,

Tanveer

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index