RE: st: RE: Hausman test for endogeneity for HT panel estimator

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Simone,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] On Behalf Of Simone Sasso
> Sent: 13 January 2014 11:20
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: RE: Hausman test for endogeneity for HT panel estimator
> 
> Dear Mark,
> Thanks a lot for your help. Is there any other way to test the endogeneity of
> the variables before running the -xthtaylor- or do i necessarily have to pick
> the variables that I expect to be endogenous beforehand, run the -xthtaylor-
> with them as -endog- and then test their endogeneity with the -xtoverid-?

The latter.  "Endogeneity" in this context means a failure of the orthogonality assumption that E(Ze)=0, where Z is your instrument and e is the disturbance in the equation you're estimating.  The disturbance e is of course unobserved, so testing this assumption in effect means working with ehat, i.e., the residual.  But you need to estimate the model to get the residual.

Cheers,
Mark

> Many thanks and have a nice day,
> Simone
> 
> 
> Il 1/13/2014 11:26 AM, Schaffer, Mark E ha scritto:
> > Simone,
> >
> > -xtoverid-, available from SSC in the usual way, provides a test of
> overidentifying restrictions after estimation by -xthtaylor-.  This may be what
> you're looking for.
> >
> > HTH,
> > mark
> >
> > *********
> >
> > From: owner-statalist@hsphsun2.harvard.edu
> > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of simone
> > sasso
> > Sent: 13 January 2014 10:00
> > To: statalist@hsphsun2.harvard.edu
> > Subject: st: Hausman test for endogeneity for HT panel estimator
> >
> > Hello,
> > I am working on a large panel dataset. I've run the Hausman test to
> compare random vs. fixed effect finding that random effect are not
> consistent. I would like to try using the Hausman and Taylor model, beliving
> that in my analysis some of the estimators are endogenous and some are
> exogenous. In order to run the Hasman test between FE and HT estimators,
> do I first have to arbitrarly assume which are the variable that are endogeous
> or how can I see that?
> > Many thanks in advance for your help,
> > Simone
> >
> >
> > -----
> > Sunday Times Scottish University of the Year 2011-2013 Top in the UK
> > for student experience Fourth university in the UK and top in Scotland
> > (National Student Survey 2012)
> >
> >
> > We invite research leaders and ambitious early career researchers to
> > join us in leading and driving research in key inter-disciplinary themes.
> > Please see www.hw.ac.uk/researchleaders for further information and
> > how to apply.
> >
> > 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/faqs/resources/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/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/


----- 
Sunday Times Scottish University of the Year 2011-2013
Top in the UK for student experience
Fourth university in the UK and top in Scotland (National Student Survey 2012)

We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.

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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/