Factorial Multivariate Bayesian Causal Forests: heterogeneous main and interaction effects of multiple treatments on correlated outcomes
Abstract
Many studies expose units to several binary treatments at once and record several correlated outcomes, yet analysts usually estimate one treatment's average effect on one outcome at a time -- discarding how treatments interact and how their effects vary across units. We introduce the factorial multivariate Bayesian causal forest, a Bayesian nonparametric model that decomposes the factorial response surface over the treatment lattice into a prognostic sum-of-trees plus one sum-of-trees per main or interaction effect, each with correlated multivariate leaf parameters, estimated jointly with coherent uncertainty. A single indicator-weighted kernel samples every component -- the prognostic term is the special case whose indicator is unity -- so the two-treatment model, the single-treatment multivariate causal forest, and a general order-r truncation are one and the same sampler. We give the ANOVA/Möbius identification of each estimand, an efficient Rcpp engine, and an interpretability layer using value-suppressing uncertainty maps. In simulations the average-effect estimators are unbiased with nominal coverage, consistent, and robust under misspecification, failing only under unmeasured confounding, which we flag. We illustrate the method on clinical, agricultural and economic data and on a deeper application to the NHANES survey. Software is provided as an R package.
Related
- Debiased and Simultaneous Inference for Heterogeneous Factorial Effect Modifiers via Residualized Walsh-Hadamard Scores
- Triple Products of Correlated Multivariate Hermite Polynomials
- Finite Mixtures of Generalized Estimating Equations for Clustering Multivariate Correlated Outcomes
- A multiscale lattice Boltzmann solver for free fluid/porous medium interaction in heterogeneous domains
- Digital Consequences of Legendre’s Formula for Factorial Valuations
BibTeX
@misc{sarti2026factorial,
title = {{Factorial Multivariate Bayesian Causal Forests: heterogeneous main and interaction effects of multiple treatments on correlated outcomes}},
author = {Danilo A. Sarti},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.31332},
eprint = {2609.31332},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.31332},
}
RIS
TY - UNPB
TI - Factorial Multivariate Bayesian Causal Forests: heterogeneous main and interaction effects of multiple treatments on correlated outcomes
AU - Danilo A. Sarti
PY - 2026
DA - 2026/09/25
PB - arXiv
DO - 10.48550/arxiv.2609.31332
UR - https://arxiv.org/abs/2609.31332
AB - Many studies expose units to several binary treatments at once and record several correlated outcomes, yet analysts usually estimate one treatment's average effect on one outcome at a time -- discarding how treatments interact and how their effects vary across units. We introduce the factorial multivariate Bayesian causal forest, a Bayesian nonparametric model that decomposes the factorial response surface over the treatment lattice into a prognostic sum-of-trees plus one sum-of-trees per main or interaction effect, each with correlated multivariate leaf parameters, estimated jointly with coherent uncertainty. A single indicator-weighted kernel samples every component -- the prognostic term is the special case whose indicator is unity -- so the two-treatment model, the single-treatment multivariate causal forest, and a general order-r truncation are one and the same sampler. We give the ANOVA/Möbius identification of each estimand, an efficient Rcpp engine, and an interpretability layer using value-suppressing uncertainty maps. In simulations the average-effect estimators are unbiased with nominal coverage, consistent, and robust under misspecification, failing only under unmeasured confounding, which we flag. We illustrate the method on clinical, agricultural and economic data and on a deeper application to the NHANES survey. Software is provided as an R package.
ER -