Logo image
Back
Robust Bayesian model averaging for linear regression models with heavy-tailed errors
Journal article   Peer reviewed

Robust Bayesian model averaging for linear regression models with heavy-tailed errors

Shamriddha De and Joyee Ghosh
Journal of applied statistics, Vol.53(2), pp.304-330
01/25/2026
DOI: 10.1080/02664763.2025.2511938

View Online

Abstract

Our goal is to develop a Bayesian model averaging technique in linear regression models that accommodates heavier tailed error densities than the normal distribution. Motivated by the use of the Huber loss function in the presence of outliers, the Bayesian Huberized lasso with hyperbolic errors has been proposed and recently implemented in the literature. Since the Huberized lasso cannot enforce regression coefficients to be exactly zero, we propose a Bayesian variable selection approach with spike and slab priors to address sparsity more effectively. The shapes of the hyperbolic and the Student-t density functions differ. Furthermore, the tails of a hyperbolic distribution are less heavy compared to those of a Cauchy distribution. Thus, we propose a flexible regression model with an error distribution encompassing both the hyperbolic and the Student-t family of distributions, with an unknown tail heaviness parameter, that is estimated based on the data. It is known that the limiting form of both the hyperbolic and the Student-t distributions is a normal distribution. We develop an efficient Gibbs sampler for posterior computation. Through simulation studies and analyzes of real datasets, we show that our method is competitive with various state-of-the-art methods.
 Generalized hyperbolic distribution huberized Bayesian lasso hyperbolic distribution Markov chain Monte Carlo model composition (MC3) spike and slab priors student-t distribution

Details

Metrics

3 Record Views
1 Times Cited - Web of Science
Logo image