Bayesian quantile regression analysis of complex survey data under informative sampling

Abstract

The interest in considering the relation among random variables in quantiles instead of the mean has emerged in various fields. Data collected from complex survey designs are of fundamental importance to different areas. Combining both frameworks provides a powerful tool for supporting decisions and is useful for practitioners from diverse backgrounds. In this thesis, we aim to advance in this literature by investigating new developments and extensions of Bayesian methods for quantile regression analysis of complex survey data under informative sampling. We not only focus on the absolutely continuous case as previous works on the topic but also develop methods for count data and multiple outputs. Our methods are particularly appealing as they provide effective and easy-to-implement methodological tools.