|
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
|
|
Keywords:
|
|
|
Abstract:
|
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
|
