Unsure how to proceed with prior and posterior predictive checking for Bayesian multiple logistic regression

Thank you so much for taking the time to reply with such thoughtful responses! I appreciate your use of questions and brainstorming to help me better understand this workflow.

I will certainly review the notebook you linked; I actually had reviewed it previously, but I guess I never made it to that last example at that bottom in enough depth…

And the last bit you provided is especially insightful: My data is actually highly imbalanced (i.e., ~90-10% split). I am going to take all of this back to the drawing board.

Have a great day!

1 Like