Department of Mathematics
Sparse sufficient dimension reduction using optimal scoring
Sufficient dimension reduction is a body of theory and methods for reducing the dimensionality of predictors while preserving information on regressions. In this paper we propose a sparse dimension reduction method to perform interpretable dimension reduction. It is designed for situations in which the number of correlated predictors is very large relative to the sample size. The new procedure is based on the optimal scoring interpretation of the sliced inverse regression method. As a result, the regression framework of optimal scoring facilitates the use of commonly used regularization techniques. Simulation studies demonstrate the effectiveness and efficiency of the proposed approach. © 2012 Elsevier B.V. All rights reserved.
High dimensionality, Linear discriminant analysis, Optimal scoring, Sliced inverse regression, Sparsity, Sufficient dimension reduction
Source Publication Title
Computational Statistics & Data Analysis
Link to Publisher's Edition
Wang, T., & Zhu, L. (2013). Sparse sufficient dimension reduction using optimal scoring. Computational Statistics & Data Analysis, 57 (1), 223-232. https://doi.org/10.1016/j.csda.2012.06.015