Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Christine Scheef <christine.scheef@unisg.ch> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | Re: st: Using ivregress when the endogenous variable is used in an interaction term in the main regression |
| Date | Fri, 23 Dec 2011 09:16:48 +0100 |
Thank you Tirthankar. You answer was very helpful! Two more questions: 1) Wooldridge states in the first step that "ALL exogenous variables" should be included to estimate the binary endogenous variable (thus, X1 in this case). So, I do not include my instruments Z in this equation? Do I have to perform any test of fit? Or does the probit/chi2 just needs to be significant? 2) Considering my third point. the instruments Z are thus, e.g. X2hat*X1. To meet the overidentification restriction I would need one more intstrument than endogenous variable. Can I include a further instrumental variable in ivregress? Best, Christine * * 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/