Hi, I am looking to run a Vuong test to compare model specification after using the -frontier command. This requires to extract individual log likelihoods. I cannot find a straight way of computing them. Does anybody have an advice about how to compute them after -frontier?

thanks,
Matías