07-09-2014 06:29 PM
I am modeling hospital visits (from a household survey where zero visits was an option and accounts for about 70%).
This is a longitudinal study (3 time points).
A zero inflated Poisson did not fit the data well.
Is there a way to calculate a two part model for longitudinal data in SAS?
07-10-2014 08:03 AM
I think NLMIXED may be your only option, and even that will be messy. I would start by doing a very careful search of the SAS-L archives (https://listserv.uga.edu/cgi-bin/wa?A0=SAS-L) for posts by David Cassell on the topic of zero-inflated and hurdle models using NLMIXED.