I have a situation where I have two binomial responses (A out of X and B out of Y). I have a binary categorical predictor C with a multilevel categorical covariate D. I basically want to find a way of doing the following regression
(A/X / B/Y) ~ C + D while ideally not losing information by simply taking the ratio of A/X and B/Y .
What regression methods are useful for this situation? (I'd prefer something with an R implementation).