|
Activity Number:
|
152
- Recent Development in Sufficient Dimension Reduction
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 31, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
International Statistical Institute
|
|
Abstract #322980
|
|
|
Title:
|
An Adaptive Approach to Dimension Reduction
|
|
Author(s):
|
Qin Wang*
|
|
Companies:
|
Virginia Commonwealth University
|
|
Keywords:
|
asymmetric least squares ;
dimension reduction ;
expectile regression ;
quantile regression
|
|
Abstract:
|
Sufficient dimension reduction provides a useful tool to high dimensional data analysis. We propose here an adaptive approach to estimate the central subspace through regression expectiles. Asymmetric least squares based nonparametric estimation is proposed to implement the new approach. The correspondence between expectiles and quantiles assures the exhaustive estimation of the central subspace. Compared to the quantile regression based dimension reduction methods, the proposed approach is computationally more efficient.
|
Authors who are presenting talks have a * after their name.
