I’m interested in using Stan to estimate probabilities for AB tests with multiple variants and I’ve been following this pymc tutorial.
I’ve managed to translate between pymc and Stan okay, but I’m reaching a problem in the Hierarchical Modeling section, where the author defines hyperpriors for the beta distribution using guidance from BDA. I can’t figure out how I might go about doing this within Stan. I’ve been looking around, but can’t find any examples for anything other than pymc.
Can anybody give me any pointers on how I might go about doing this?