Dirichlet process mixtures of block $g$ priors for model selection and prediction in linear models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Porwal, Anupreet, Rodriguez, Abel
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914555981987840
author Porwal, Anupreet
Rodriguez, Abel
author_facet Porwal, Anupreet
Rodriguez, Abel
contents This paper introduces Dirichlet process mixtures of block $g$ priors for model selection and prediction in linear models. These priors are extensions of traditional mixtures of $g$ priors that allow for differential shrinkage for various (data-selected) blocks of parameters while fully accounting for the predictors' correlation structure, providing a bridge between the literatures on model selection and continuous shrinkage priors. We show that Dirichlet process mixtures of block $g$ priors are consistent in various senses and, in particular, that they avoid the conditional Lindley ``paradox'' highlighted by Som et al. (2016). Further, we develop a Markov chain Monte Carlo algorithm for posterior inference that requires only minimal ad-hoc tuning. Finally, we investigate the empirical performance of the prior in various real and simulated datasets. In the presence of a small number of very large effects, Dirichlet process mixtures of block $g$ priors lead to higher power for detecting smaller but significant effects without only a minimal increase in the number of false discoveries.
format Preprint
id arxiv_https___arxiv_org_abs_2411_00471
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Dirichlet process mixtures of block $g$ priors for model selection and prediction in linear models
Porwal, Anupreet
Rodriguez, Abel
Methodology
Machine Learning
This paper introduces Dirichlet process mixtures of block $g$ priors for model selection and prediction in linear models. These priors are extensions of traditional mixtures of $g$ priors that allow for differential shrinkage for various (data-selected) blocks of parameters while fully accounting for the predictors' correlation structure, providing a bridge between the literatures on model selection and continuous shrinkage priors. We show that Dirichlet process mixtures of block $g$ priors are consistent in various senses and, in particular, that they avoid the conditional Lindley ``paradox'' highlighted by Som et al. (2016). Further, we develop a Markov chain Monte Carlo algorithm for posterior inference that requires only minimal ad-hoc tuning. Finally, we investigate the empirical performance of the prior in various real and simulated datasets. In the presence of a small number of very large effects, Dirichlet process mixtures of block $g$ priors lead to higher power for detecting smaller but significant effects without only a minimal increase in the number of false discoveries.
title Dirichlet process mixtures of block $g$ priors for model selection and prediction in linear models
topic Methodology
Machine Learning
url https://arxiv.org/abs/2411.00471