Abstract: We develop a Bayesian "sum-of-trees" model where each tree is constrained by a regularization prior to be a weak learner, and fitting and inference are accomplished via an iterative Bayesian backfitting MCMC algorithm that generates samples from a posterior. Effectively, BART is a nonparametric Bayesian regression approach which uses dimensionally adaptive random basis elements. Motivated by ensemble methods in general, and boosting algorithms in particular, BART is defined by a statistical model: a prior and a likelihood. This approach enables full posterior inference including point and interval estimates of the unknown regression function as well as the marginal effects of potential predictors. By keeping track of predictor inclusion frequencies, BART can also be used for model-free variable selection. BART's many features are illustrated with a bake-off against competing methods on 42 different data sets, with a simulation experiment and on a drug discovery classification problem.
Recommendations
- Bayesian additive regression trees with model trees
- Log-linear Bayesian additive regression trees for multinomial logistic and count regression models
- Bayesian regression trees for high-dimensional prediction and variable selection
- Variable selection for BART: an application to gene regulation
- Bayesian treed models
Cites work
- A Bayesian CART algorithm
- A decision-theoretic generalization of on-line learning and an application to boosting
- A spatially-adjusted Bayesian additive regression tree model to merge two datasets
- Alternative models for stock price dynamics.
- An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation Bias
- Bagging predictors
- Bayesian Analysis of Binary and Polychotomous Response Data
- Bayesian backfitting. (With comments and a rejoinder).
- Bayesian Inference on Network Traffic Using Link Count Data
- Greedy function approximation: A gradient boosting machine.
- scientific article; zbMATH DE number 527274 (Why is no real title available?)
- Least angle regression. (With discussion)
- Multivariate adaptive regression splines
- Reversible jump Markov chain Monte Carlo computation and Bayesian model determination
- Statistical Analysis of Financial Data in S-Plus
- The Bayesian additive classification tree applied to credit risk modelling
Cited in
(only showing first 100 items - show all)- Experimental Evaluation of Individualized Treatment Rules
- Bayesian non-parametric ordinal regression under a monotonicity constraint
- Learning certifiably optimal rule lists for categorical data
- A flexible sensitivity analysis approach for unmeasured confounding with multiple treatments and a binary outcome with application to SEER-medicare lung cancer data
- borrowr
- Estimating treatment effect heterogeneity in randomized program evaluation
- Logitboost autoregressive networks
- Simultaneous dimension reduction and variable selection in modeling high dimensional data
- Modeling threshold interaction effects through the logistic classification trunk
- Bayesian additive regression trees using Bayesian model averaging
- Tree ensembles with rule structured horseshoe regularization
- Energy bagging tree
- Bayesian multiple response kernel regression model for high dimensional data and its practical applications in near infrared spectroscopy
- Adaptive-modal Bayesian nonparametric regression
- Variable selection and sensitivity analysis using dynamic trees, with an application to computer code performance tuning
- BayesTree
- Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (with discussion)
- Bayesian additive regression trees with model trees
- Tree based credible set estimation
- Toward an explainable machine learning model for claim frequency: a use case in car insurance pricing with telematics data
- Data driven conditional optimal transport
- Uncertainty quantification for honest regression trees
- Uncertainty quantification for Bayesian CART
- The dependent Dirichlet process and related models
- Analyzing stochastic computer models: a review with opportunities
- Monotonic effects of characteristics on returns
- Bayesian data synthesis and the utility-risk trade-off for mixed epidemiological data
- Robust discrete choice models with \(t\)-distributed kernel errors
- Nowcasting in a pandemic using non-parametric mixed frequency VARs
- Optional Pólya trees: posterior rates and uncertainty quantification
- Model-guided adaptive sampling for Bayesian model selection
- Inference in Bayesian additive vector autoregressive tree models
- Spatial CART classification trees
- Rule-based Bayesian regression
- Robust Bayesian inference for big data: combining sensor-based records with traditional survey data
- Augmented direct learning for conditional average treatment effect estimation with double robustness
- Heterogeneous causal effects with imperfect compliance: a Bayesian machine learning approach
- BART with targeted smoothing: an analysis of patient-specific stillbirth risk
- Estimating the health effects of environmental mixtures using Bayesian semiparametric regression and sparsity inducing priors
- Data-driven modelling of the Reynolds stress tensor using random forests with invariance
- Nonparametric Bayesian analysis of the compound Poisson prior for support boundary recovery
- Posterior concentration for Bayesian regression trees and forests
- Minimax optimal rates for Mondrian trees and forests
- A decision-theoretic approach for model interpretability in Bayesian framework
- Inflection points in community-level homeless rates
- Targeted smooth Bayesian causal forests: an analysis of heterogeneous treatment effects for simultaneous vs. interval medical abortion regimens over gestation
- A semiparametric modeling approach using Bayesian additive regression trees with an application to evaluate heterogeneous treatment effects
- Predicting missing values: a comparative study on non-parametric approaches for imputation
- Variance prior forms for high-dimensional Bayesian variable selection
- Maximum relevance minimum common redundancy feature selection for nonlinear data
- Neural random forests
- Variable prioritization in nonlinear black box methods: a genetic association case study
- Learning algorithms to evaluate forensic glass evidence
- Estimating population average causal effects in the presence of non-overlap: the effect of natural gas compressor station exposure on cancer mortality
- Automated versus do-it-yourself methods for causal inference: lessons learned from a data analysis competition
- Comment: Contributions of model features to BART causal inference performance using ACIC 2016 competition data
- Minimax-optimal nonparametric regression in high dimensions
- A Bayesian regression tree approach to identify the effect of nanoparticles' properties on toxicity profiles
- Multiple imputation for sharing precise geographies in public use data
- Bayesian nonparametric regression with varying residual density
- Machine learning applied to asteroid dynamics
- Smoothing and adaptation of shifted Pólya tree ensembles
- Flexible tree-structured regression models for discrete event times
- Do forecasts of bankruptcy cause bankruptcy? A machine learning sensitivity analysis
- Entity resolution with empirically motivated priors
- Biomarker-driven adaptive design
- Sequential design for computer experiments with a flexible Bayesian additive model
- Efficient Bayesian multivariate surface regression
- Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks
- Estimating Individual Treatment Effect in Observational Data Using Random Forest Methods
- A Generalized Estimating Equation Approach to Multivariate Adaptive Regression Splines
- Heteroscedastic BART via Multiplicative Regression Trees
- A Semiparametric Bayesian Approach to Dropout in Longitudinal Studies With Auxiliary Covariates
- Scalable Bayesian Nonparametric Clustering and Classification
- Bayesian Deep Net GLM and GLMM
- Cluster-specific variable selection for product partition models
- dbarts
- An adaptive sampling scheme guided by BART -- with an application to predict processor performance
- Tree models of Bayesian regression: a case study
- Assessing lack of common support in causal inference using Bayesian nonparametrics: Implications for evaluating the effect of breastfeeding on children's cognitive outcomes
- Influencing elections with statistics: targeting voters with logistic regression trees
- A Bayesian CART algorithm
- Bayesian neural networks for selection of drug sensitive genes
- Estimation and Inference of Heterogeneous Treatment Effects using Random Forests
- A Bayesian machine learning approach for optimizing dynamic treatment regimes
- Bayesian regression tree ensembles that adapt to smoothness and sparsity
- Variable selection for BART: an application to gene regulation
- High-dimensional Bayesian inference in nonparametric additive models
- Spike-and-slab priors for function selection in structured additive regression models
- Embedding black-box regression techniques into hierarchical Bayesian models
- Bayesian regression trees for high-dimensional prediction and variable selection
- Model guided adaptive design and analysis in computer experiment
- FLAME: a fast large-scale almost matching exactly approach to causal inference
- scientific article; zbMATH DE number 7370554 (Why is no real title available?)
- Log-linear Bayesian additive regression trees for multinomial logistic and count regression models
- Do German economic research institutes publish efficient growth and inflation forecasts? A Bayesian analysis
- Measuring the algorithmic convergence of randomized ensembles: the regression setting
- Bayesian additive regression trees in spatial data analysis with sparse observations
- scientific article; zbMATH DE number 7625170 (Why is no real title available?)
- Comparing Covariate Prioritization via Matching to Machine Learning Methods for Causal Inference Using Five Empirical Applications
This page was built for publication: BART: Bayesian additive regression trees
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q65651)
