Online Program Home
My Program

Abstract Details

Activity Number: 381 - High-Dimensional Nonparametric Statistics
Type: Invited
Date/Time: Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
Sponsor: Section on Nonparametric Statistics
Abstract #326708 Presentation
Title: Robust Estimation, Efficiency, and Lasso Debiasing
Author(s): Po-Ling Loh*
Companies: UW-Madison

We present results concerning high-dimensional robust estimation for linear regression with non-Gaussian errors. We provide error bounds for certain local/global optima of penalized M-estimators, valid even when the loss function employed is nonconvex -- giving rise to more robust estimation procedures. We also present a new approach for robust location/scale estimation with rigorous theoretical guarantees. We conclude by discussing high-dimensional variants of one-step estimation procedures from classical robust statistics and connections to recent work on confidence intervals based on Lasso debiasing.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program