|
Activity Number:
|
345
- High-Dimensional Statistics
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #306939
|
Presentation
|
|
Title:
|
Debiased Inference in High-Dimensional Single-Index Models Under Gaussian Design
|
|
Author(s):
|
Hamid Eftekhari* and Moulinath of Banerjee and Ya'acov Ritov
|
|
Companies:
|
University of Michigan and university of michigan and university of michigan
|
|
Keywords:
|
Sparsity;
Asymptotic normality;
Non-linear link function;
Semiparametric regression;
Average Partial Effect
|
|
Abstract:
|
We consider the problem of statistical inference of a single covariate in a single-index model with p > n covariates and unknown link function under Gaussian design. The estimator of the coefficient is similar to the de-biased lasso in the standard linear model and is square-root consistent and asymptotically normally distributed.
|
Authors who are presenting talks have a * after their name.
