Bayesian nonparametric (BNP) methods can be used to flexibly model joint or conditional distributions, as well as functional relationships. These methods, along with causal assumptions, can be used with the g-formula for inference about causal effects. This general approach to causal inference has several possible advantages over popular semiparametric methods, including efficiency gains, the ease of causal inference on any functionals of the distribution of potential outcomes, the use of prior information, and capturing uncertainty about causal assumption via informative prior distributions. In this short course we review BNP methods and illustrate their use for causal inference in the setting of point treatments, dynamic (longitudinal) treatments, and mediation. We present several data examples and discuss software implementation using R. The R code and/or packages used to run the data examples will be provided to the attendees at a specific github site.
Instructor(s): Michael Daniels, University of Florida; Jason Roy, University of Pennsylvania