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? Any help will be appreciated very much. Sharon |
Sharon,
> -----Original Message----- > From: [hidden email] > [mailto:[hidden email]] On Behalf Of > xueliansharon > Sent: 21 July 2010 02:08 > To: [hidden email] > 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 > -- > View this message in context: > http://statalist.1588530.n2.nabble.com/overidentification-test > -and-marginal-effects-after-ivprobit-tp5319208p5319208.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/ > -- 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/ |
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 |
<> 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 -----Ursprüngliche Nachricht----- Von: [hidden email] [mailto:[hidden email]] Im Auftrag von xueliansharon Gesendet: Donnerstag, 22. Juli 2010 04:47 An: [hidden email] 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/ |
Sharon,
> -----Original Message----- > From: [hidden email] > [mailto:[hidden email]] On Behalf Of > Martin Weiss > Sent: 22 July 2010 08:08 > To: [hidden email] > 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: '[hidden email]' > Subject: RE: overidentification test and marginal effects > after ivprobit > > Sharon, > > > -----Original Message----- > > From: [hidden email] > > [mailto:[hidden email]] On Behalf Of > > xueliansharon > > Sent: 21 July 2010 02:08 > > To: [hidden email] > > 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: [hidden email] > [mailto:[hidden email]] Im Auftrag von > xueliansharon > Gesendet: Donnerstag, 22. Juli 2010 04:47 > An: [hidden email] > 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 > 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/ |
Hi, Martin and Mark,
Thanks for all your help! And sorry for the inconvenience I caused. I will be more careful about quoting next time. Best, Sharon |
Free forum by Nabble | Edit this page |