A0609
Title: Multivariate response directional regression: An approach via projective resampling method
Authors: Ah Reum Lee - Ewha Womans University (Korea, South) [presenting]
Abstract: Directional regression is a widely used tool to implement sufficient linear dimension reduction, which is useful for extracting core information from a high-dimensional dataset. The integration of a projective resampling method with sufficient dimension reduction, as outlined by another study, facilitates the recovery of dimension reduction subspaces in multivariate response regression. A novel approach is presented to multivariate response dimension reduction by combining projective resampling with directional regression. The proposed method is demonstrated to offer a practical alternative to existing methods. Through both simulation studies and real data analysis, the method is illustrated to outperform other dimension reduction methods in a number of scenarios.
