Hello all:
I have a two stage model. The first stage is a selection equation with a categorical variable (druginsr):1. no insurance, 2. government sponsored insurance 3. private insurance. A set of social economic status and chronic conditions are included as explanatory variables.
And the second stage outcome is a binary variable (avoid_care) =1, if patient avoiding care because of cost, =0 otherwise. The insurance status (druginsr), a subset of social characteristics and all the chronic conditions are included as explanatory variables.

cmp (avoid_care = age sex post_educ....... druginsr) (druginsr = age sex post_educ immigrant oth_language......... ), ind(druginsr*$cmp_probit $cmp_mprobit) qui

I tried "cmp" like above and cannot get any estimates? If someone knows a different way of doing this or how to improve please please let me know.
Thank you!