Activity Number:
|
181
- SPEED: Statistical Learning and Data Science Speed Session 1, Part 2
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 29, 2019 : 10:30 AM to 11:15 AM
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
Abstract #307530
|
|
Title:
|
Inference for Measurement Error Model Under High-Dimensional Settings
|
Author(s):
|
Mengyan Li* and Yanyuan Ma
|
Companies:
|
Penn State University and The Pennsylvania State University
|
Keywords:
|
High dimensional inference;
Error in variables;
Nuisance Parameter;
Sparsity
|
Abstract:
|
We consider the problem of inference for low dimensional components with additive measurement errors in high dimensional settings. We use the consistent CoCoLasso estimator proposed by Datta et al. (2017) as the initial estimator, and generalize the decorrelated score method proposed by Ning et al. (2017) to high dimensional measurement error model. We construct a novel test statistic and provide asymptotic guarantees on the type I errors and local powers. Further, we construct asymptotically normal point estimators using 'corrected' decorrelated score function, then the corresponding confidence region can be obtained.
|
Authors who are presenting talks have a * after their name.