Bayesian empirical likelihood for quantile regression

TL;DR

This paper introduces a Bayesian empirical likelihood approach for quantile regression, enabling efficient inference and leveraging priors to improve estimates across quantiles, especially in data-sparse regions.

Contribution

It develops a Bayesian empirical likelihood framework for quantile regression, incorporating informative priors to enhance efficiency and explore commonality across quantiles.

Findings

Asymptotic normality of the posterior is established.

Empirical results show substantial efficiency gains with informative priors.

The method avoids complex likelihood maximization via MCMC.

Abstract

Bayesian inference provides a flexible way of combining data with prior information. However, quantile regression is not equipped with a parametric likelihood, and therefore, Bayesian inference for quantile regression demands careful investigation. This paper considers the Bayesian empirical likelihood approach to quantile regression. Taking the empirical likelihood into a Bayesian framework, we show that the resultant posterior from any fixed prior is asymptotically normal; its mean shrinks toward the true parameter values, and its variance approaches that of the maximum empirical likelihood estimator. A more interesting case can be made for the Bayesian empirical likelihood when informative priors are used to explore commonality across quantiles. Regression quantiles that are computed separately at each percentile level tend to be highly variable in the data sparse areas (e.g., high or low percentile levels). Through empirical likelihood, the proposed method enables us to explore various forms of commonality across quantiles for efficiency gains. By using an MCMC algorithm in the computation, we avoid the daunting task of directly maximizing empirical likelihood. The finite sample performance of the proposed method is investigated empirically, where substantial efficiency gains are demonstrated with informative priors on common features across several percentile levels. A theoretical framework of shrinking priors is used in the paper to better understand the power of the proposed method.