mBART: multidimensional monotone BART

DOI10.1214/21-BA1259arXiv1612.01619OpenAlexW3154559866MaRDI QIDQ6202919FDOQ6202919

Authors: Hugh Chipman, Edward I. George, Robert E. McCulloch, Thomas S. Shively 

Publication date: 27 February 2024

Published in: Bayesian Analysis (Search for Journal in Brave)

Abstract: For the discovery of regression relationships between Y and a large set of p potential predictors x 1 , . . . , x p , the flexible nonparametric nature of BART (Bayesian Additive Regression Trees) allows for a much richer set of possibilities than restrictive parametric approaches. However, subject matter considerations sometimes warrant a minimal assumption of monotonicity in at least some of the predictors. For such contexts, we introduce mBART, a constrained version of BART that can flexibly incorporate monotonicity in any predesignated subset of predictors using a multivariate basis of monotone trees, while avoiding the further confines of a full parametric form. For such monotone relationships, mBART provides (i) function estimates that are smoother and more interpretable, (ii) better out-of-sample predictive performance, and (iii) less post-data uncertainty. While many key aspects of the unconstrained BART model carry over directly to mBART, the introduction of monotonicity constraints necessitates a fundamental rethinking of how the model is implemented. In particular, the original BART Markov Chain Monte Carlo algorithm relied on a conditional conjugacy that is no longer available in a monotonically constrained space. Various simulated and real examples demonstrate the wide ranging potential of mBART.

Full work available at URL: https://arxiv.org/abs/1612.01619

zbMATH Keywords

Bayesian nonparametricsisotonic regressionMCMC algorithmshape constrained inferenceensemble modelmultidimensional nonparametric regression

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric regression and quantile regression (62G08)

Cites Work

• Variable selection for BART: an application to gene regulation
• BART: Bayesian additive regression trees
• Generalized Additive and Index Models with Shape Constraints
• Improving point and interval estimators of monotone functions by rearrangement
• Title not available (Why is that?)
• A method for Bayesian monotonic multiple regression
• Estimating a smooth monotone regression function
• Bayesian monotone regression using Gaussian process projection
• Bayesian Isotonic Regression and Trend Analysis
• Estimating Smooth Monotone Functions
• Title not available (Why is that?)
• A Bayesian approach to non-parametric monotone function estimation
• Bayesian Spectral Analysis Regression with Shape Restrictions
• Nonparametric function estimation subject to monotonicity, convexity and other shape constraints
• A nonparametric Bayes method for isotonic regression
• Bayesian estimation and inference for generalised partial linear models using shape-restricted splines
• Bayesian Multivariate Isotonic Regression Splines
• Concentration rate and consistency of the posterior distribution for selected priors under monotonicity constraints
• Posterior concentration for Bayesian regression trees and forests
• Monotone Smoothing with Application to Dose-Response Curve
• Bayesian Regression Tree Ensembles that Adapt to Smoothness and Sparsity
• Estimating shape constrained functions using Gaussian processes
• Correcting an estimator of a multivariate monotone function with isotonic regression

This page was built for publication: mBART: multidimensional monotone BART

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6202919)