Help with calculating Marginal Probability of each outcome using CMP

[Ganesh AR]

I have built a multivariate binary Probit model with seven binary outcomes and allowing for error correlations across these outcomes. Below is a sample code

cmp (y1 = x1) (y2 = x2) (y3 = x3) (y4 = x4) (y5 = x5) (y6 = x6) (y7 = x7), ind($cmp_probit $cmp_probit $cmp_probit $cmp_probit $cmp_probit $cmp_probit $cmp_probit)

I am now trying to predict the marginal probability of each outcome using the below code (showing only 2 outcomes for sample purposes)

predict p1, eq(y1) pr
predict p2, eq(y2) pr

I would like to know if the above code provides the marginal probability of each outcome because when I use it to recalculate the log-likelihood it matches the corresponding sum of the log-likelihood of the independent models. I would like to calculate the marginal probability of each outcome. How do I do it?

Any directions on the above is much appreciated. Thanks.