A conditional density estimation partition model using logistic Gaussian processes

DOI10.1093/BIOMET/ASZ064zbMATH Open1435.62290arXiv1703.06978OpenAlexW2992455837WikidataQ101121311 ScholiaQ101121311MaRDI QIDQ5222227FDOQ5222227

Authors: Richard D. Payne, Nilabja Guha, Yu Ding, Bani Mallick 

Publication date: 1 April 2020

Published in: Biometrika (Search for Journal in Brave)

Abstract: Conditional density estimation (density regression) estimates the distribution of a response variable y conditional on covariates x. Utilizing a partition model framework, a conditional density estimation method is proposed using logistic Gaussian processes. The partition is created using a Voronoi tessellation and is learned from the data using a reversible jump Markov chain Monte Carlo algorithm. The Markov chain Monte Carlo algorithm is made possible through a Laplace approximation on the latent variables of the logistic Gaussian process model. This approximation marginalizes the parameters in each partition element, allowing an efficient search of the posterior distribution of the tessellation. The method has desirable consistency properties. In simulation and applications, the model successfully estimates the partition structure and conditional distribution of y.

Full work available at URL: https://arxiv.org/abs/1703.06978

Recommendations

• Partition-based conditional density estimation
• Adaptive Bayesian estimation of conditional densities
• Recursive partitioning and multi-scale modeling on conditional densities
• Multivariate density estimation by Bayesian sequential partitioning
• Bayesian density regression with logistic Gaussian process and subspace projection

zbMATH Keywords

Laplace approximationreversible jump Markov chain Monte CarloBayesian conditional density estimationpartition modellogistic Gaussian process

Mathematics Subject Classification ID

Density estimation (62G07) Applications of statistics in engineering and industry; control charts (62P30) Gaussian processes (60G15) Generalized linear models (logistic models) (62J12)

Cited In (8)

• Adaptive Bayesian Sum of Trees Model for Covariate-Dependent Spectral Analysis
• Partition-based conditional density estimation
• Partition Weighted Approach For Estimating the Marginal Posterior Density With Applications
• A Nonstationary Soft Partitioned Gaussian Process Model via Random Spanning Trees
• Conditionally conjugate mean-field variational Bayes for logistic models
• Bayesian clustering of spatial functional data with application to a human mobility study during COVID-19
• Title not available (Why is that?)
• Recursive partitioning and multi-scale modeling on conditional densities

This page was built for publication: A conditional density estimation partition model using logistic Gaussian processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5222227)