st: Marginal effects after heckman

[Fred Dzanku <fdzanku@gmail.com>]

Dear All,

How can I get marginal effects of the (probit) selection equation
after running a heckman selection model by maximum likelihood? I
estimated a model in the form



                        svy: heckman y $Xi, sel($Xi $Xsel)



where y is the dependent variable; $Xi is a list of explanatory
variables, which also appear in the selection equation; and $Xsel is a
list of exclusion restrictions. I know the coefficients of the
variables in the outcome equation that also appear in the selection
equation must be adjusted in order for them to be interpreted as
marginal effects but I also want to get the marginal effects of the
selection equation. I tried the following



                         margins, dydx(*) atmeans
predict(equation(select)) vce(unconditional)



but what I get as marginal effects are exactly the same as the
coefficients of the selection equation. I've searched the web but have
no answers yet. Can you please help or do you know how I can get some
help?





--
Fred M. Dzanku
Dept. of Ag.& Food Econ.
University of Reading
PO Box 237
Reading
RG6 6AR

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/