About bartCause
bartCause is an R package that uses Bayesian Additive Regression Trees (BART) to adjust for confounding variables without making parametric assumptions. Instead of using methods like propensity score matching or weighting, it models the outcome variable through a flexible machine learning model. If we can appropriately model the outcome, we can impute missing counterfactual outcomes and then find our causal estimates. thinkCausal uses BART for causal inference, taking advantage of its non-parametric, flexible approach to outcome modeling.
While bartCause does not make any parametric assumptions, it does require that all confounding variables have been measured and included in the model.