st: A question about -margins-

[Angelica Calle <angelica.calle.s@gmail.com>]

Dear StataListers:

I am trying to estimate the probability that a estudent gets a
scholarship conditional on factors like parents' education (among
other variables). To this aim I am interested in obtaining the
marginal effects of the parents' education  on the probability of
obtaining the scholarship. I have been using the command -heckprob- to
get this estimates (my dependent variable is dichotomus), and the
command -margins, dydx(*) predict(pcond)- to obtain the marginal
effects. I know that this calculation (estimating the marginal
effects) must consider the parameters of both equations (selection and
outcome equation) and I would like to be sure that the command
-margins, dydx(*) predict(pcond)- does that. I have been searching if
this is the case or if I have to use something else, unfortunately I
was not able to identify if margins considers both equations. I would
really appreciate if you guys could give me some reference to have
certainty on this issue.

Thanks,

Angelica.
*
*   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/