st: Multivariate probit postestimation

[Susan Steiner <steiner@giga-hamburg.de>]

Dear Statalist users,

We are employing a multivariate probit model for the estimation of three 
alternative outcomes (use of savings, use of credits, use of insurance). 
We are using the commands by Capillari and Jenkins but do not know how 
to produce marginal joint probabilities (and significance levels) for 
combinations of these alternatives. Capillari and Jenkins provide the 
commands for producing marginal probabilities for each equation and for 
joint marginal probabilities for all successes and all failures (mvpred, 
pall) but not for other combinations of the alternatives (for example, 
use of savings and insurance but no use of credit). Does anybody have 
experience with doing bootstrap procedures in this case?
Thanks!
Susan

*
*   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/