Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Eleonora Paesen <eleonora.paesen@googlemail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Issue re. heckprob |
Date | Wed, 4 Jan 2012 14:11:50 +0000 |
Hello, I am investigating the transitions to parenthood after age 30 for those in the survey who are childless at age 30. I would like to take into account for the selection of couples who are still childless at age 30, by applying a heckprob model. Since I want to estimate time to entry into parenthood, I have a probit model as a selection model (first-step equation) and a discrete time survival model for the latter binary variable as a behavioural equation (second step). Can the Heckprob command deal with this - or is it set up for a dataset that has only one line per person i.e. that is wide? If so, can we do the Heckprob physically in two parts - i.e. get the inverse mills ratio from the selection model and then put this into a discrete time probit model? Thanks. Best, Eleonora * * 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/