Probit Monotone BART
This work addresses a specific limitation in nonparametric modeling for binary outcomes, representing an incremental advancement in the BART framework.
The authors tackled the problem of estimating monotonic functions for binary outcomes by extending monotone BART to handle binary data, resulting in a method called probit monotone BART that enables conditional mean function estimation in such cases.
Bayesian Additive Regression Trees (BART) of Chipman et al. (2010) has proven to be a powerful tool for nonparametric modeling and prediction. Monotone BART (Chipman et al., 2022) is a recent development that allows BART to be more precise in estimating monotonic functions. We further these developments by proposing probit monotone BART, which allows the monotone BART framework to estimate conditional mean functions when the outcome variable is binary.