The estimating equations for quantile regression (QR) can be solved using an EM algorithm in which the M-step is computed via weighted least squares, with weights computed at the E-step as the expectation of independent generalized inverse-Gaussian variables. The package extends QR to allow for random effects in the linear predictor. The generalized alternating minimization (GAM) framework also allows for variable selection in the QR, large P setting.
For more details on the theory, see the paper on ArXiV. For more usage documentation, see the QREM.pdf file in the doc/ subfolder in this repository .