I want to analyze the determinants of credit constraints of a firm. I have information for both formal and informal credit. I have 6 categories of credit-constraint statuses of a firm for formal credit, while I only have 2 categories for informal credit. As credit constraint status is determined simultaneously, and formal and informal credit market are interrelated, I want to estimate the determinants of credit constraints for formal and informal credit market at the same time. Bivariate logit/probit model may be too simplistic. Is there extension of bivariate logit model or some kind of multivariate multinomial logit model in Stata? Thank you very much in advance.
$\begingroup$ Have you looked at a latent class model? Basically, each decision-maker has an estimated probability of being given one of several choice sets. The coefficients for each set, and the coefficients for the latent class model, are all determined simultaneously. $\endgroup$– gregmacfarlaneOct 2, 2012 at 23:11
$\begingroup$ Thank you for your reply. I am not sure but my dependent variables are not latent variables, and the aim is to identify the determinants of credit-constraint statuses (defined by two variables - one is binary, and the other is 6 category variables). Sorry if my explanation is not clear. $\endgroup$– TanoOct 3, 2012 at 16:09