|
Activity Number:
|
464
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #308743 |
|
Title:
|
Shrinkage Inverse Regression for Model-Free Variable Selection
|
|
Author(s):
|
Lexin Li*+ and Howard D. Bondell
|
|
Companies:
|
North Carolina State University and North Carolina State University
|
|
Address:
|
Department of Statistics, Raleigh, NC, 27695,
|
|
Keywords:
|
Inverse regression estimation ; Nonnegative garrote ; Sufficient dimension reduction ; Variable selection
|
|
Abstract:
|
The family of inverse regression estimators recently proposed by Cook and Ni (2005) have proven effective in dimension reduction by transforming the high-dimensional predictor vector to its low-dimensional projections. In this talk, we propose a general shrinkage estimation strategy for the entire inverse regression estimation family, which is capable of simultaneous dimension reduction and variable selection. We demonstrate that the new estimators achieve consistency in variable selection without requiring any traditional model, meanwhile retaining the root-n estimation consistency of the dimension reduction basis. We also show the effectiveness of the new estimators through both simulation and real data analysis.
|
