I've run negative binomial hurdle models and chose one based off of AIC and BIC.
I'm having trouble figuring out how to validate my model. K-fold test doesn't work because the classifications (more than 2) are not balanced. I can't figure out how to do stratified K-folds in R (lots of code for Python online). I've looked into creating stratified folds using caret but I'm not advanced enough to understand what to do afterwards. I also tried looking into likelihood ratio test and it seems to be way beyond my level.
I'd really appreciate any help figuring out how to finish the whole modeling process.
