From ff736eac62d352450326ea302db5bb95096b02e5 Mon Sep 17 00:00:00 2001 From: Drew Herren Date: Wed, 17 Dec 2025 13:51:21 -0600 Subject: [PATCH] Include stochtree under Bayesian approaches for HTE estimation --- CausalInference.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/CausalInference.md b/CausalInference.md index 954ef82..3ab2980 100644 --- a/CausalInference.md +++ b/CausalInference.md @@ -187,9 +187,10 @@ generate balance tables and plots before and after covariate balancing. `r pkg(" Some of the above mentioned packages can also be used for heterogeneous treatment effect (HTE) estimation. -- *Bayesian approaches* for individual causal effect estimation are +- *Bayesian approaches* for heterogeneous causal effect estimation are available in `r pkg("bartCause")` (based on Bayesian Additive - Regression Trees). + Regression Trees) and `r pkg("stochtree")` (which implements the Bayesian + Causal Forest method, based on Bayesian Additive Regression Trees). - *Fisherian approaches* for an omnibus test of heterogeneity and decomposition of overall treatment effect heterogeneity into a systematic component explained by covariates and an idiosyncratic component is implemented in `r pkg("hettx")`. - *Efficacious treatment or population subset selection* exploiting treatment effect heterogeneity is implemented in