I was trying to compute the effects of a variable in the treatment equation
on the y-variables, and I was
wondering if there a better way to do it than just by hand.
Specifically, consider
treatreg y x w, treat(z = u w)
and I would like to compute
dy/dw = y_2 + y_3 z_2, (*)
where y_2 is the direct effect of w on y, y_3 if the effect of z on y, and
z_2 is how the probability that z =1 changes when w changes.
I can use mfx to compute y_2, y_3, z_2 and then use (*) to compute dy/dw,
but I was wondering
if there a better way of doing this.
Richard Boylan
Associate Professor of Economics
University of Alabama
Culverhouse College of Commerce & Business Administration
Tuscaloosa, AL 35487-002
205-348-8967, fax 205-348-0590
*
* 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/