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

The analysis of longitudinal data poses a series of issues, but it also gives the chance to observe changes in the unit behavior over time which may be of prime interest. This has been the focus of a huge literature in the context of linear and generalized linear regression which, in the last ten years or so, has moved to the context of linear quantile regression models for continuous responses. In this paper, we present lqmix, a novel R package that helps estimate a class of linear quantile regression models for longitudinal data, in the presence of time-constant and/or time-varying, unit-specific, random coefficients, having unspecific distribution. Model parameters are estimated in a maximum likelihood framework, via an extended EM algorithm, and parameters' standard errors are estimated via a block-bootstrap procedure. The analysis of a benchmark dataset is used to give details of the package functions.