Sorry, folks, the last email trucated the first line of what I sent out.
Here's my question again:
I'd like to know if anyone has done any work on programming in a
seemingly unrelated negative binomial model, along the lines of those
proposed by Winkelmann (2000). On a related topic, can the ml command
in stata accomodate multi-equation models?
F. Michael Kunz, M.A.
Senior Research Associate
Center for Health Economic and Policy Studies
Medical University of South Carolina
Department of Health Administration and Policy
19 Hagood Avenue, 408 Harborview Tower
P.O. Box 250807
Charleston, South Carolina 29425