lqmix: an R package for longitudinal data analysis via linear quantile mixtures

TL;DR

lqmix is an R package designed for analyzing longitudinal data using linear quantile regression models with random coefficients, employing an EM algorithm for estimation and bootstrap for standard errors.

Contribution

The paper introduces lqmix, a novel R package that enables flexible linear quantile regression for longitudinal data with random effects, expanding existing methodologies.

Findings

Successfully applied to a benchmark dataset.

Provides estimation of models with unspecified random coefficient distributions.

Includes functions for comprehensive data analysis and inference.

Abstract

The analysis of longitudinal data gives the chance to observe how unit behaviors change over time, but it also poses a series of issues. These have been the focus of an extensive literature in the context of linear and generalized linear regression moving also, in the last ten years or so, to the context of linear quantile regression for continuous responses. In this paper, we present \texttt{lqmix}, a novel \texttt{R} package that assists in estimating a class of linear quantile regression models for longitudinal data, in the presence of time-constant and/or time-varying, unit-specific, random coefficients, with unspecified distribution. Model parameters are estimated in a maximum likelihood framework via an extended EM algorithm, while parameters' standard errors are derived via a block-bootstrap procedure. The analysis of a benchmark dataset is used to give details on the package functions.