st: treatreg with second stage as a probit?

["Stephen V. Burks" <svburks@morris.umn.edu>]

Hi, All. (excuse duplication if it occurs, but I think I forgot to 
make the first version plain text)

Can anyone suggest a simple way to run treatreg when the second stage 
regression is also a probit (or better, a 3-level multinomial or 
ordered qualtitative response variable model)?

My colleagues and I have done this manually (i.e. create inverse 
Mills ratio from a first-stage selection probit and use it as a 
regressor in the second stage treatment probit), but suspect the 
usual standard errors are then incorrect.

(FYI, we thought about a more general model, but our N is too small 
to make separate Heckman models run on each subset of the two 
"selected" types very helpful--assuming this would be the correct 
more general model for our situation.)

Thanks,
Steve

--------------------------------------------------
*Stephen V. Burks, Ph.D.
*
*Associate Professor of Economics and Management
*University of Minnesota, Morris
*
*Division of Social Sciences
*600 East 4th Street
*Morris, MN 56267-2134
*
*Voice: (320) 589-6191
*Division Secretary: (320) 589-6200
*FAX: (320) 589-6117
*E-Mail:  svburks@mrs.umn.edu
*
*Striving to be the best national
*public liberal arts college in the US.
------------------------------------------------


--------------------------------------------------
*Stephen V. Burks, Ph.D.
*
*Associate Professor of Economics and Management
*University of Minnesota, Morris
*
*Division of Social Sciences
*600 East 4th Street
*Morris, MN 56267-2134
*
*Voice: (320) 589-6191
*Division Secretary: (320) 589-6200
*FAX: (320) 589-6117
*E-Mail:  svburks@mrs.umn.edu
*
*Striving to be the best national
*public liberal arts college in the US.
------------------------------------------------

*
*   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/