bartcs-package {bartcs} | R Documentation |
bartcs: Bayesian Additive Regression Trees for Confounder Selection
Description
Fit Bayesian Regression Additive Trees (BART) models to select true confounders from a large set of potential confounders and to estimate average treatment effect. For more information, see Kim et al. (2023) doi:10.1111/biom.13833.
Details
Functions in bartcs
serve one of three purposes.
Functions for fitting:
separate_bart()
andsingle_bart()
.Functions for summary:
summary()
andplot()
.Utility function for OpenMP:
count_omp_thread()
.
The code of BART model are based on the 'BART' package by
Sparapani et al. (2021) under the GPL license, with modifications.
The modifications from the BART
package include (but are not limited to):
Add CHANGE step.
Add Single and Separate Model.
Add causal effect estimation.
Add confounder selection.
References
Sparapani R, Spanbauer C, McCulloch R (2021). “Nonparametric Machine Learning and Efficient Computation with Bayesian Additive Regression Trees: The BART R Package.” Journal of Statistical Software, 97(1), 1–66. doi:10.18637/jss.v097.i01
Kim, C., Tec, M., & Zigler, C. M. (2023). Bayesian Nonparametric Adjustment of Confounding, Biometrics doi:10.1111/biom.13833