Create your own community (if you need to)

Back in 1991 I went to a conference of Bayesians and I was disappointed that the vast majority seem to not be interested in checking their statistical models. The attitude seemed to be, first, that model checking was not possible in a Bayesian context, and, second, that model checking was illegitimate because models were subjective. No wonder Bayesianism was analogized to a religion.

This all frustrated me, as I’d found model checking to be highly relevant in my Bayesian research in two different research problems, one involving inference for emission tomography (which had various challenges arising from spatial models and positivity constraints), the other involving models for district-level election results.

The good news is that, in the years since our book Bayesian Data Analysis came out, a Bayesian community has developed that is more accepting of checking models by looking at their fit to data. Many challenges remain.

The point of this story is that sometimes you can work with an existing community, sometimes you have to create your own community, and sometimes it’s a mix. In this case, my colleagues and I did not try to create a community on our own; we very clearly piggybacked off the existing Bayesian community, which indeed included lots of people who were interested in checking model fit, once it became clear that this was a theoretically valid step.

P.S. For more on the theoretical status of model checking in Bayesian inference, see this 2003 paper, A Bayesian formulation of exploratory data analysis and goodness-of-fit testing and this 2018 paper, Visualization in Bayesian workflow.

P.P.S. Zad’s cat, pictured above, is doing just fine. He doesn’t need to create his own community.