st: Marginal effects of log-linear sample selection models

[Shehzad Ali <sia500@york.ac.uk>]

Dear statalisters,

I need some help with calculating marginal effects of dummy variables in a
log-linear model after adjusting for sample selection and endogeneity
biases. I would truly appreciate any help on this.

In my model, the final outcome variable is log_y which is only observed if
selection_var=1. Both log_y and selection_var are also affected by a common
endogenous variable (endo). selection_var and endo are both dummies.

The model I am using is summarised below:

biprobit (selection_var= x1 x2 endo) (endo=x1 x2 x3)

Then I predict Inverse Mills Ratio (imr) for selection_var, followed by:

treatreg log_y x1 x2 imr, treat(endo=x1 x2 x3)

This gives me a coefficient for endo but I dont know how this would be 
interpreted because endo also indirectly appears through IMRs. Can you 
kindly suggest how to estimate the marginal effect of endo on y? Will mfx 
compute, dyex give me the correct interpretation?

Sincere thanks,

Ali


--
Shehzad I Ali
Department of Social Policy & Social Work
University of York
YO10 5NG
+44 (0) 773-813-0094

*
*   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/