B1206
Title: Dimension reduction with expectation of a conditional difference measure
Authors: Wenhui Sheng - Marquette University (United States) [presenting]
Abstract: A flexible model-free approach is introduced to sufficient dimension reduction analysis using the expectation of a conditional difference measure. Without any strict conditions, such as linearity condition or constant covariance condition, the method estimates the central subspace effectively under linear or nonlinear relationships between response and predictors. The method is especially useful when the response is categorical. We also studied the root-n consistency and asymptotic normality properties of the estimates. The efficacy of our method is demonstrated through both simulations and real data analysis.
