PyMC3 posterior prediction example; how to incorporate a matrix of betas?

Hi,

I would recommend using InferenceData so that xarray takes care of all broadcasting automatically. There are some examples of this in the rugby analytics example and in the multilevel model example. This second example has examples of xarray usage but with the posterior variables, not with posterior predictive. Once you have generated the inferencedata, there are no practical differences in operating with posterior, prior or posterior predictive.

1 Like