st: RE: AW: RE: overidentification test and marginal effects after ivprobit

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

Sharon,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Martin Weiss
> Sent: 22 July 2010 08:08
> To: statalist@hsphsun2.harvard.edu
> Subject: st: AW: RE: overidentification test and marginal 
> effects after ivprobit
> 
> 
> <> 
> 
> Sharon, the FAQ do recommend that you edit the text you are 
> replying to 
> (http://www.stata.com/support/faqs/res/statalist.html#others),
>  but to totally delete it makes it very difficult to put your 
> contribution in perspective. Maybe quote the core of what you 
> are replying to above your own text? 
> 
> 
> 
> HTH
> Martin

Martin is right (as usual).  To speed things up, here is the original exchange plus a comment:

> -----Original Message-----
> From: Schaffer, Mark E 
> Sent: 21 July 2010 18:49
> To: 'statalist@hsphsun2.harvard.edu'
> Subject: RE: overidentification test and marginal effects 
> after ivprobit
> 
> Sharon,
> 
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu
> > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> > xueliansharon
> > Sent: 21 July 2010 02:08
> > To: statalist@hsphsun2.harvard.edu
> > Subject: st: overidentification test and marginal effects after 
> > ivprobit
> > 
> > Dear all,
> > 
> > I want to estimate an ivprobit model, and I also want to 
> conduct the 
> > overidentification test and want to compute marginal effects after 
> > estimating the model.  Now the problem is if I use 
> -overid-, I need to 
> > estimate ivprobit with "twostep" option, but if I use "twostep" 
> > estimation, mfx command can't be used to compute the 
> marginal effects 
> > since option "predict(p)" is not available for twostep estimator of 
> > ivprobit. So what should I do to do both overidentification 
> test and 
> > marginal effects after ivprobit?
> 
> If memory serves, the overid test result for the two-step 
> estimator carries over to the ML estimator because they use 
> the same exclusion restrictions.  (You might want to double 
> check this though - don't rely on my memory!)  If so, you can 
> just use the two-step overid test statistic as your 
> overidentification test for your ML estimation.
> 
> --Mark
> 
> > Any help will be appreciated very much.
> > 
> > Sharon

And now the answer to Sharon's question is easier to understand: when I said "the overid test result for the two-step estimator", I meant simply the test statistic that overid reports after ivprobit with the twostep option.

--Mark

> 
> 
> -----Ursprüngliche Nachricht-----
> Von: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von 
> xueliansharon
> Gesendet: Donnerstag, 22. Juli 2010 04:47
> An: statalist@hsphsun2.harvard.edu
> Betreff: st: RE: overidentification test and marginal effects 
> after ivprobit
> 
> Hi, Mark,
> 
> Thanks for your response. But I don't understand what " the 
> two-step overid test statistic" is, could you please 
> illustrate it more explicitly?
> 
> Many thanks,
> Sharon
> --
> View this message in context:
> http://statalist.1588530.n2.nabble.com/overidentification-test
-and-marginal-
> effects-after-ivprobit-tp5319208p5323802.html
> Sent from the Statalist mailing list archive at Nabble.com.
> *
> *   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/