A1044
Title: Estimating heterogeneous causal mediation effects with bayesian decision tree ensembles
Authors: Antonio Linero - University of Texas at Austin (United States) [presenting]
Angela Ting - University of Texas at Austin (United States)
Abstract: The causal inference literature has increasingly recognized that explicitly targeting treatment effect heterogeneity can lead to improved scientific understanding and policy recommendations. Towards the same end, studying the causal pathway connecting the treatment to the outcome can also be useful. These problems are addressed in the context of causal mediation analysis. A varying coefficient model based on Bayesian additive regression trees is introduced to identify and regularize heterogeneous causal mediation effects; analogously with linear structural equation models, these effects correspond to covariate-dependent products of coefficients. It is shown that, even on large datasets with few covariates, LSEMs can produce highly unstable estimates of the conditional average direct and indirect effects, while the Bayesian causal mediation forests model produces stable estimates. It is found that the approach is conservative, with effect estimates "shrunk towards homogeneity". The method's salient properties are examined using data from the Medical Expenditure Panel Survey and empirically grounded simulated data. Finally, the purpose is to show how the model can be combined with posterior summarization strategies to identify interesting subgroups and interpret the model fit.
