Online Program Home
My Program

Abstract Details

Activity Number: 136 - Recent Advances in Dimension Reduction
Type: Contributed
Date/Time: Monday, July 29, 2019 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Learning and Data Science
Abstract #305144
Title: A Sufficient Dimension Reduction Method via Expectation of Conditional Difference
Author(s): Qingcong Yuan* and Wenhui Sheng and Xiangrong Yin
Companies: Miami University and Marquette University and University of Kentucky
Keywords: Sufficient Dimension Reduction; Central Subspace

Using an expectation of conditional difference measure, we introduce a novel approach to sufficient dimension reduction problems. The proposed method is model-free and is especially useful when the response is categorical. The estimation of dimension and the central subspace using the measure is discussed and the dimension reduction for large p and small n cases is developed. The root-n consistency and asymptotic normality is established under regularity conditions. Numerical studies are provided to demonstrate the advantage of the method. The proposed method is very competitive and robust comparing to existing dimension reduction methods.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program