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

Re: st: Inverse Mills


From   "Renzo Comolli" <renzo.comolli@yale.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Inverse Mills
Date   Thu, 25 Sep 2003 22:23:45 -0400

Hi Danny

>I have the following model.
>Y1 = b0 + a1Y1 + a3X3 + e1 (Y1 is binary)
>Y2 = b1 + a2Y2 + a4X4 + e2 (Y2 is observed but its truncated at 0)

I really think there must still be one typo. I should read
Y1 = b0 + a1Y2 + a3X3 + e1 (Y1 is binary)
Y2 = b1 + a2Y1 + a4X4 + e2 (Y2 is observed but its truncated at 0)


I undestand the intuitive idea you are shooting for. I have never seen what
you propose. 
You have to check the validity of your procedure.
Clearly you have the poblem that when you do one regression and compute the
inverse mills ration you have the other endogenous variable in, and that
does not bode well. But I am not sure, you have to check it.

Good news now!!
The technique that I would use to solve your problem (assuming your problem
is in the way I wrote it) is just a modification of the technique which is
reprogrammed in -reg3-
. help reg3

Study the logic of -reg3-, if you have the manuals read it in there. If you
don't have the manuals read 3SLS (3 stages least squares on text books),
then instead of using -reg3- do the 3 steps using the estimators that you
need to take into account the nature of Y1 and Y2

More good news!!
As you know probit is preprogrammed
. help probit

Also tobit is preprogrammed!!!!!!!!!
. help tobit

I don't guarantee that the procedure I propose here, is a correct and valid
one.  That's what I would do if I had really little time to think about the
problem.

Good luck.

Ciao
Renzo

*
*   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 1996–2022 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index