I am trying to use the bayesglm function from the arm package to deal with the complete separation problem in my data as detailed previously in this thread How to deal with perfect separation in logistic regression?. My question is how to compare nested models (used to use anova when using non Bayesian logistic regression) to come up with the best model from a set of variables. Furthermore, need to know how to evaluate the model fit. Many thanks in advance
Asked
Active
Viewed 66 times
