Title: Penalized Weighted Least Squares for Outlier Detection and Robust Regression

Abstract: To conduct regression analysis for data contaminated with outliers, many approaches have been proposed for simultaneous outlier detection and robust regression, so is the approach proposed in this manuscript. This new approach is called "penalized weighted least squares" (PWLS). By assigning each observation an individual weight and incorporating a lasso-type penalty on the log-transformation of the weight vector, the PWLS is able to perform outlier detection and robust regression simultaneously. A Bayesian point-of-view of the PWLS is provided, and it is showed that the PWLS can be seen as an example of M-estimation. Two methods are developed for selecting the tuning parameter in the PWLS. The performance of the PWLS is demonstrated via simulations and real applications.