时间:2019年4月12日13:00-14:00

地点:浦东校区第五教学楼224室

主题:Generalized Kernel-based Inverse Regression Methods for Sufficient Dimension Reduction

主讲人:谢传龙

主讲人简介：谢传龙博士于2017年香港浸会大学数学系毕业，现在在暨南大学经济学院统计系工作。主要研究方向为模型检验，充分降维，测量误差模型。

报告内容简介：The linearity condition and the constant conditional variance assumption popularly used in sufficient dimension reduction are respectively close to elliptical symmetry and normality. These conditions are often required by inverse regression-based methods. However, it is always the concern about their restrictiveness. In this paper, we give systematic studies to provide insight into the reasons why the popularly used sliced inverse regression and sliced average variance estimation need these conditions. Then we propose a new framework to relax these conditions and suggest generalized kernel-based inverse regression methods to handle a class of mixture multivariate skew elliptical distributions. Numerical studies show that generalized kernel inverse regression (GKIR) and generalized kernel average variance estimation (GKAVE) work well when the mentioned two conditions do not satisfy for the distributions in this class.