Dear Austin,
Thanks very much for your response. I also thought to use xtoverid,
but unfortunately the procedure that I must use, xtpmg (which employs
pooled mean-group estimator � la Pesaran, Shin, and Smith (1999) for
non-stationary heterogenous panel data sets) does not support it.
Hence, I am stuck!
Sincerely,
Mark
On Tue, Aug 5, 2008 at 7:01 PM, Austin Nichols <[email protected]> wrote:
> Mark--
> Try using -xtoverid-
> (view http://fmwww.bc.edu/RePEc/bocode/x/xtoverid.hlp
> for a preview):
>
> ssc inst xtoverid, replace
> webuse abdata
> xtreg n w k if year>=1978 & year<=1982, re
> xtoverid
> di r(j)
> est store re
> xtreg n w k if year>=1978 & year<=1982, fe
> est store fe
> hausman fe re, sigmaless
> di r(chi2)
>
>
> On 8/5/08, Mark Dincecco <[email protected]> wrote:
>> Dear all,
>>
>> I am running xtpmg regressions for mg and pmg. The problem is that
>> when I run (hausman mg pmg, sigmamore), even using the (sigmamore)
>> option, a negative test statistic is returned.
>>
>> In a 1984 Econometric article (p. 1226), Hausman and McFadden
>> interpret a negative H score as strong evidence of failure to reject
>> the null hypothesis that the difference in coefficient values is not
>> systematic.
>>
>> My question is: Is there anything else I can or should try to do to
>> produce a positive H statistic? Before I take the negative H score as
>> evidence that IIA holds, I would like to be sure that I am not missing
>> something obvious.
>>
>> Thanks very much. Your suggestions are be greatly appreciated.
>>
>> Sincerely,
>> Mark
> *
> * 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/
