I would like to know if there is an underlying reason to choose twopm over Craggs hurdle model or vice-versa. I have data on biweekly expenditures on medication with a greater part of my observations recording zero expenditures. I have tried the stata commands for Craggs double hurdle (probit & truncreg, craggit, dhurdle, churdle) and there seems to be an issue with convergence. twopm works fine and I'm likely to go ahead and use twopm but I wanted to be sure if there is a reason to use one over the other.
-
Login or Register
- Log in with
