TY - DATA T1 - A Parallel Algorithm for Large-Scale Nonconvex Penalized Quantile Regression PY - 2017/10/09 AU - Liqun Yu AU - Nan Lin AU - Lan Wang UR - https://tandf.figshare.com/articles/journal_contribution/A_Parallel_Algorithm_for_Large-scale_Nonconvex_Penalized_Quantile_Regression/5024429 DO - 10.6084/m9.figshare.5024429.v2 L4 - https://ndownloader.figshare.com/files/8477783 KW - ADMM KW - Nonconvex penalty KW - Parallelization KW - Quantile regression and single-loop algorithm N2 - Penalized quantile regression (PQR) provides a useful tool for analyzing high-dimensional data with heterogeneity. However, its computation is challenging due to the nonsmoothness and (sometimes) the nonconvexity of the objective function. An iterative coordinate descent algorithm (QICD) was recently proposed to solve PQR with nonconvex penalty. The QICD significantly improves the computational speed but requires a double-loop. In this article, we propose an alternative algorithm based on the alternating direction method of multiplier (ADMM). By writing the PQR into a special ADMM form, we can solve the iterations exactly without using coordinate descent. This results in a new single-loop algorithm, which we refer to as the QPADM algorithm. The QPADM demonstrates favorable performance in both computational speed and statistical accuracy, particularly when the sample size n and/or the number of features p are large. Supplementary material for this article is available online. ER -