21 Feb 2022, noon (Central time):

Some Bayesian Methods for Causal Inference

Rather than offering any new conceptual framework, we simply discuss several different areas where Bayesian inference can make a difference in causal inference. These areas include learning from noisy experiments, generalization to new scenarios, meta-analysis, decision analysis, and models for measurement error. We discuss applications in biology, psychology, marketing, and other areas. These ideas should be relevant if you are interested in using Bayesian methods and also if you want to use these ideas to understand other statistical approaches. If there is time, we will also discuss open questions, challenges, and places where we’re stuck.

They invited me to speak and I said sure and asked what they wanted to hear about, and they said: “We would like you to talk about practical uses of Bayesian methods in observational studies, particularly on problems with the identification of weak effects.” The above is what I came up with. We’ll see how it goes!