Empirical Bayesian ordinal regression model

I can only recommend what @bbbales2 said in the other thread. I hope I don’t sound harsh here, but I think the problems are a bit more conceptual. I know that Bayesian reasoning is often sold as updating prior beliefs after seeing data, then looking at new data with the new prior and update again and so on. But this is not what is typically done Bayesian Statistics. At least I don’t think it is helpful to think about it that way. I guess that is what @bbbales2 was trying to convey (just estimate your data jointly!).

I’d recommend you pick up a good textbook and familiarize yourself a bit more with this approach to Bayesian modeling. I always recommend the Gelman and Hill textbook or Statistical Rethinking by Richard McElreath.

1 Like