I have the following setup:
I have an event C_EVENT that takes binary values (with 1 being event taking
place).
There are aset of variables caleed VARS that characterize this event.
I need two things:
1) for each event, I want the predicted probit event probability (EVENTPROB)
2) the inverse mills ratio, IMR, to carry out second pass linear
regressions.
My series of commands are:
probit C_EVENT VARS
predict EVENTPROB /* gets the predicted
probabilities*/
predict PROBITXB, xb
gen PDFPROBIT = normd(PROBITXB)
gen CDFPROBIT = normprob(PROBITXB)
gen IMR = PDFPROBIT/CDFPROBIT /* gets the inverse mills ratio*/
reg NEWVAR IMR REMAINVAR /* second pass regression*/
Am I doing it correctly?
Or should I use:
heckman NEWVAR REMAINVAR, select (VARS) twostep
predict EVENTPROB
Feedback will be greatly appreciated.
Thanks,
Subhankar Nayak
Assistant Professor (Finance)
DuPree College of Management
Georgia Institute of Technology
Atlanta, GA 30332
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
