Marco,
-hausman- isn't valid with -robust- (see Hayashi, "Econometric", 2000, p. 234, note 18).
Steve's advice to use -xtoverid- is the right way to go. It implements a Hausman-like GMM test that extends to robust VCV estimators.
--Mark
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Steven Archambault
> Sent: 13 August 2009 16:24
> To: [email protected]
> Subject: Re: st: hausman test
>
> The positive definite error is concerning. I would suggest
> using xtoverid as an alternative test.
>
>
> Schaffer, M.E., Stillman, S. 2006. xtoverid: Stata
> module to calculate tests of overidentifying restrictions
> after xtreg, xtivreg,
> xtivreg2 and xthtaylor
> http://ideas.repec.org/c/boc/bocode/s456779.html
>
>
>
>
> On Thu, Aug 13, 2009 at 8:38 AM, Marco
> Buur<[email protected]> wrote:
> > I run hausman test and got Prob>chi2 = 0.1668 According my
> > undestanding this is confmation that I should use random
> effect model.
> > Am i right ?
> >
> > Thanks a lot
> >
> > xtreg A B C , fe robust
> > est store fixed
> > xtreg A B C , re robust
> > hausman fixed
> >
> >
> > Test: Ho: difference in coefficients not systematic
> >
> > chi2(8) = (b-B)'[(V_b-V_B)^(-1)](b-B)
> > = 11.67
> > Prob>chi2 = 0.1668
> > (V_b-V_B is not positive definite)
> >
> >
> >
> >
> > xtreg Public_cap_ln Prevalence_ln GNI_cap_ln reg4 reg8 Year_* if
> > Tag!=1 , fe robust
> >
> > est store fixed
> >
> > xtreg Public_cap_ln Prevalence_ln GNI_cap_ln reg4 reg8 Year_* if
> > Tag!=1 , re robust
> >
> > hausman fixed
> > *
> > * 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/
>
--
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.
*
* 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/
