11-30-2016 05:19 PM
I am conducting a matched case control study with 1:m matching with a multinomial outcome. I know I can conduct conditional logitic regression in proc logistic with the strata statement and I can conduct a multinomal regression with the link=glogit statement. However, they seem to not be compatible with each other in proc logistic. Is there a way to do this analysis in SAS? Any help would be greatly appreciated.
12-01-2016 04:58 PM
The conditional model isn't possible, but you can model clustered, multinomial data using PROC GEE or using PROC SURVEYLOGISTIC with a CLUSTER statement. Both allow you to fit the model without having to estimate cluster-specific parameters.