Bayesian projection pursuit regression

DOI10.1007/S11222-023-10334-ZMaRDI QIDQ6190647zbMATH OpenOpenAlexFDO

Authors Gavin Q. Collins, Devin Francom, Kellin N. Rumsey

Publication date 6 February 2024

Published in Statistics and Computing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2210.09181

zbMATH Keywords

Bayesian nonparametric statisticscomputer experimentsemulatorspline basisuncertainty quantification

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Nonparametric regression and quantile regression (62G08)

Abstract: In projection pursuit regression (PPR), an unknown response function is approximated by the sum of M "ridge functions," which are flexible functions of one-dimensional projections of a multivariate input space. Traditionally, optimization routines are used to estimate the projection directions and ridge functions via a sequential algorithm, and M is typically chosen via cross-validation. We introduce the first Bayesian version of PPR, which has the benefit of accurate uncertainty quantification. To learn the projection directions and ridge functions, we apply novel adaptations of methods used for the single ridge function case (M=1), called the Single Index Model, for which Bayesian implementations do exist; then use reversible jump MCMC to learn the number of ridge functions M. We evaluate the predictive ability of our model in 20 simulation scenarios and for 23 real datasets, in a bake-off against an array of state-of-the-art regression methods. Its effective performance indicates that Bayesian Projection Pursuit Regression is a valuable addition to the existing regression toolbox.

Recommendations

• Generalized projection pursuit regression
• Tests in projection pursuit regression
• Estimation of projection pursuit regression via alternating linearization
• Projection pursuit regression for moderate nonlinearities
• Moderate projection pursuit regression for multivariate response data

Cites work

• A Gaussian process regression approach to a single-index model
• A Projection Pursuit Algorithm for Exploratory Data Analysis
• BART: Bayesian additive regression trees
• Bayesian backfitting. (With comments and a rejoinder).
• Bayesian estimation and variable selection for single index models
• Calculation Methods for Wallenius' Noncentral Hypergeometric Distribution
• Greedy function approximation: A gradient boosting machine.
• scientific article; zbMATH DE number 3716572 (Why is no real title available?)
• scientific article; zbMATH DE number 1522714 (Why is no real title available?)
• scientific article; zbMATH DE number 2148835 (Why is no real title available?)
• scientific article; zbMATH DE number 845714 (Why is no real title available?)
• Inferring atmospheric release characteristics in a large computer experiment using Bayesian adaptive splines
• Mixtures of g Priors for Bayesian Variable Selection
• Multidimensional Additive Spline Approximation
• Multivariate adaptive regression splines
• Random forests
• Rank-normalization, folding, and localization: an improved \(\widehat{R}\) for assessing convergence of MCMC (with Discussion)
• Reversible jump Markov chain Monte Carlo computation and Bayesian model determination
• Selected Works of David Brillinger
• Sensitivity analysis and emulation for functional data using Bayesian adaptive splines
• The identification of a particular nonlinear time series system
• Visualizing the Effects of Predictor Variables in Black Box Supervised Learning Models

Cited in

(4)

• Generalized Bayesian MARS: tools for stochastic computer model emulation
• Projection pursuit Bayesian regression for symmetric matrix predictors
• Co-Active Subspace Methods for the Joint Analysis of Adjacent Computer Models
• Elastic Bayesian model calibration

This page was built for publication: Bayesian projection pursuit regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6190647)