Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

re: st: mprobit with mills ratio?


From   Ian Watson <i.watson@econ.usyd.edu.au>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   re: st: mprobit with mills ratio?
Date   Tue, 4 Oct 2005 14:05:01 +1000

Manisha Shah <shah@are.Berkeley.EDU> asked:

MS> Does anyone know how to generate/define a Mills ratio from the
MS> estimation of a multinomial probit (with 3 choices)? 
MS>  
MS> Thanks!

MS> *
MS> *   For searches and help try:
MS> *   http://www.stata.com/support/faqs/res/findit.html
MS> *   http://www.stata.com/support/statalist/faq
MS> *   http://www.ats.ucla.edu/stat/stata/


The process is the same as for a probit model, except you need to
specify which outcome is the relevant one for your prediction. You may
need to repeat this twice, if you want two (inverse) mills ratios.

For example, if working has three outcomes (work full-time, work
part-time, not work), and you want an inverse mills ratio for the
working part-time outcome:

mprobit work  age edu children region
capture drop phat
capture drop wmills
predict phat if e(sample), xb outcome(2)
gen wmills = normden(phat)/norm(phat)


Then just repeat this for working full-time by specifying outcome(1)
and using different variable names for hat and mills.
  

-- 
Kind regards, 
Ian

-------------------------------
Ian Watson
Senior Researcher
acirrt, University of Sydney
NSW, 2006, Australia

phone: 02 9351 5622
email:i.watson@econ.usyd.edu.au
www.acirrt.com
-------------------------------

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index