國立政治大學統計學系
學 術 演 講
主講人：黃世豪副教授(國立中央大學數學系)學 術 演 講
題 目：Conditional Independence Testing for General Sufficient Dimension
Reduction Methods
時 間：民國114年10月20日 (星期一) 下午1：30
地 點：國立政治大學逸仙樓050101教室
摘 要：
We study conditional independence testing within the sufficient dimension reduction (SDR) framework. The goal is to assess whether selected predictors contribute to explaining the response after controlling for the others, with SDR alleviating the curse of dimensionality and preserving modeling flexibility. We propose a novel procedure that performs conditional independence testing by combining appropriate residualization with SDR dimension testing. The procedure is adaptable to a broad class of SDR methods, allowing the direct application of existing dimension tests. Simulations show our procedure achieves empirical performance comparable or superior to that of existing methods in several settings. (This is joint work with Dr. Hsin-wen Chang and Dr. Kerby Shedden.)
Keywords: Conditional independence test, Coordinate test, Dimension test, Residualization, Sufficient dimension reduction
