04-04-2014 10:52 AM
I am using proc genmod to fit a logistic regression model with repeated measures. The data are at the person-year level (one row per each person in every year). Each person-year row has time-relevant variables regarding subjects and a time-relevant outcome (0/1). I am using the following code:
proc genmod data=r.data descending ;
class var1 var2 id;
model outcomevar= var1 var2 var3/dist=bin
The model works fine, what I need to know is how to produce odds ratio estimates instead of the normal genmod output. I have some categorical variables, some binomial, and some continuous.
Any help is much appreciated.