光华讲坛——社会名流与企业家论坛第6589期

主题：Nonlinear dependence metrics and their statistical applications 非线性相依度量及其统计应用（系列讲座）

主讲人：伊利诺伊大学厄巴纳-香槟分校 邵晓峰教授

主持人：西南财经大学统计学院 常晋源教授

时间：6月26日 8:30-12:00  6月27日 8:30-12:00  6月28日 8:30-12:00

举办地点：西南财经大学光华校区光华裙楼2303

主办单位：数据科学与商业智能联合实验室 统计学院 科研处

主讲人简介：

Xiaofeng Shao received his PhD in Statistics from the University of Chicago in 2006 and has since been a faculty member with the Department of Statistics at the University of Illinois Urbana-Champaign. His current research interests include time series analysis, change-point analysis, functional data analysis, high dimensional data analysis and their applications. He is a fellow of Institute of Mathematical Statistics (IMS) and American Statistical Association (ASA). He currently serves as an associate editor for Journal of Royal Statistical Society, Series B and Journal of Time Series Analysis.

邵晓峰于2006年在芝加哥大学获得统计学博士学位，此后一直在伊利诺伊大学厄巴纳-香槟分校统计系任教。他目前的研究兴趣包括时间序列分析、变点分析、函数型数据分析、高维数据分析及其应用。他是当选的ASA和IMS成员，目前担任Journal of Royal Statistical Society, Series B和Journal of Time Series Analysis的副主编。

内容简介：

In these talks, we aim to provide an introduction of nonlinear dependence metrics and their statistical applications. Emphasis will be placed on distance covariance, energy distance and their variants, including Hilbert-Schmidt Independence Criterion, maximum mean discrepancy, martingale difference divergence, among others. The usefulness of these metrics will be demonstrated in some contemporary problems in statistics, such as dependence testing and variable screening/selection for high-dimensional data, as well as dimension reduction and diagnostic checking for multivariate time series. Some recent work on their applications to the inference of non-Euclidean data will also be discussed. The presentations are based on the research results I have obtained in the past and will cover methodology, theory and practical data examples.

在本系列讲座中，我们旨在介绍非线性相依度量及其统计应用，将重点讲距离协方差、能量距离及其变体，包括希尔伯特-施密特独立准则、最大均值差异、鞅差发散等。这些度量将在统计学的一些当代问题中得到应用，如高维数据的相依度量和变量筛选/选择，以及多变量时间序列的降维和检测。此外，还将讨论将这些指标应用于非欧几里得数据推断的一些最新工作。以上内容是基于我过去取得的研究成果，将涵盖方法论、理论和实际数据示例。