Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Sara Velezmoro <sv330@cam.ac.uk> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: 2SLS with oprobit first-stage - obtaining fitted values? |
Date | Tue, 04 Feb 2014 00:02:17 +0100 |
Hi,I am trying to instrument an ordinal categorical variable (with five categories) by using oprobit in the first stage of the 2SLS procedure.
When I use the fitted probabilities from this in the second stage I obtain uninterpretable coefficient estimates (they add up to 1), so a colleague suggested I use fitted values instead.
So my question is: how do I obtain fitted values after oprobit using STATA?
(More details: my dependent variable is log(wages), and I am regressing this on father's education (which is split into five categories); by using fitted probabilities from oprobit in the second stage, the coefficient estimates on the dummies add up to 1, which I cannot interpret in this context. I am hoping that the solution lies in using fitted values instead!)
Many thanks for your help! Sara * * 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/