Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Heckman-type selection and cmp


From   Chris Smith <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: Heckman-type selection and cmp
Date   Tue, 7 Jan 2014 14:40:34 +0000

Dear all,
I wish to run a panel data Heckman-type correction model in which the selection equation would be multinomial probit and the outcome equation is continuous. The selection equation would be run on the full sample and would segment the sample into eight groups (including the zero group). The idea is that a separate outcome equation would be run for each of the seven non-zero groups in the second stage. 
I am aware that cmp is the appropriate command for this type of model but I am struggling to think how best to specify the ind() option for the outcome equation. I need to tell cmp to run a separate continuous regression for each of the seven non-zero groups that result from the selection model. I suspect that simply specifying $cmp_cont would result in all non-zero groups being lumped together but I may be wrong. If I am correct, does anyone have any ideas about how I might specify the ind() option?
For completeness, the type of model that I wish to run is below and the element of the model that I am uncertain about is indicated by ???.
cmp (FirmPerf = Xs || firm_ID: ) (OUTCOME_GROUP = Xs || firm_ID: ), ind(???*$cmp_cont $cmp_mprobit) redraws (#)
Many thanks,
Chris 		 	   		  
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index