统计与数据科学系系列学术报告之四百一十五-四百一十七期

统计与数据科学系系列学术报告之四百一十六期

 

时   间:2023年12月8日(周五)14:00-15:00

地   点:史带楼501室

主持人:复旦大学 管理学院 统计与数据科学系 张新生 教授

报告人:Professor  Heping Zhang   Susan Dwight Bliss Professor of Biostatistics, Department of Statistics and Data Science, Yale University

   目:Tensor quantile regression with applications in brain imaging data analysis

摘   要:Neuroimaging studies through analysis of magnetic resonance imaging (MRI) data are important in our understanding of brain function. A critical technique in those studies is tensor regression models in which a scalar outcome is regressed against an array of images collectively called tensor. The high dimensionality of the tensor makes it necessary to reduce the dimension of the data through decomposition of the data as well as constrained optimalization by making use of the imaging data structures. The non-normality of the response also calls for attention to the use of quantile regression techniques.  In this present, we will discuss relevant statistical techniques in tensor quantile regression and an application in brain imaging data analysis.

个人简介:Dr. Zhang published over 370 research articles and monographs in theory and applications of statistical methods and in several areas of biomedical research including epidemiology, genetics, child and women health, mental health, substance use, and reproductive medicine. He directed a training program in mental health research that was funded by the NIMH. He directs the Collaborative Center for Statistics in Science that coordinates the Reproductive Medicine Network to evaluate treatment effectiveness for infertility. He is a fellow of the American Statistical Association and a fellow of the Institute of Mathematical Statistics. He was named the 2008 Myrto Lefokopoulou distinguished lecturer by Harvard School of Public Health and a Medallion Lecturer by the Institute of Mathematical Statistics. In 2011, he received the Royan International Award on Reproductive Health. Dr. Zhang was the president of the International Chinese Statistical Association in 2019. He is a former-editor of the Journal of the American Statistical Association - Applications and Case Studies. He was the recipient of the 2022 Neyman Award and Lecture by the Institute of Mathematical Statistics and the 2023 Distinguished Achievement Award by the International Chinese Statistical Association.

 

统计与数据科学系系列学术报告之四百一十六期

 

时    间:2023年12月8日(周五)15:00-16:00

地    点:史带楼501室

主持人:复旦大学 管理学院 统计与数据科学系 朱仲义 教授

报告人:王学钦 教授  中国科学技术大学

题   目:Tree Building via Hypothesis Tests

摘   要:Tree building, often viewed as an algorithmic modeling technique, is rapidly expanding and has become a crucial component of machine learning in the last decade. Due to its unique algorithmic characteristics, the study of its statistical properties from the standpoint of statistical hypothesis testing has been limited. In this work, a tree is built by recursively testing whether or not nodes are split. We establish the asymptotic distribution of testing and illustrate how Type 1 and Type 2 errors spread across the tree structure. This framework demonstrates how to conduct hypothesis testing in a recursive structure and permits statistical inferences about the tree’s complexity. Moreover, we demonstrate the effectiveness of our strategy through several numerical studies.

个人简介:中国科学技术大学管理学院教授, 2003年毕业于纽约州立大学宾汉姆顿分校(Binghamton University)。目前担任教育部高等学校统计学类专业教学指导委员会委员、统计学国际期刊《JASA》、《SII》、《CJS》的Associate Editor、高等教育出版社《Lecture Notes: Data Science, Statistics and Probability》系列丛书的副主编、中国现场统计研究会数据科学与人工智能分会副理事长、中国青年统计学家协会副会长、全国工业统计学教学研究会数字经济与区块链技术协会副理事长、中国工业统计学教学研究会常务理事和中国统计教育学会常务理事等。

目前主要从事统计机器学习的理论和方法研究,研究方向还包括现代统计推断(针对非欧数据的推断和递归推断)、精准医疗、统计优化和计算、风险管理和政策评估等。

 

统计与数据科学系系列学术报告之四百一十七期

 

时    间:2023年12月8日(周五)16:00-17:00

地    点:史带楼501室

主持人:复旦大学 管理学院 统计与数据科学系 朱仲义 教授

报告人:朱文圣 教授  东北师范大学

题   目:Robust Covariate Balancing Method in Learning Optimal Individualized Treatment Regimes

摘   要:Personalized medicine has recently received increasing attention since patients have heterogeneous responses to treatment. An important part of personalized medicine is individualized treatment regime (ITR), which helps medical practitioners to provide more precise treatment. That is, it can be designed to recommend treatment decisions to patients based on their individual characteristics and to maximize the overall clinical benefit to the patient. Most of the existing statistical methods usually assume an outcome regression model or a propensity score model to construct the value function. However, if any of the above assumptions are invalid, the estimated treatment regime is not reliable. In this article, we first define a contrast value function, which is the basis of the study for ITR. Then we construct a general framework of a hybrid estimator to estimate the contrast value function by combining two types of estimation methods. We further propose a covariate balancing robust (CBR) estimator of the contrast value function by combining the inverse probability weighted (IPW) method and matching method, which is based on Covariate Balancing Propensity Score (CBPS) proposed by Imai and Ratkovic (2014). The theoretical results show that the CBR estimator is doubly robust, that is, it is consistent if either the propensity score model or the matching is correct. Through a large number of simulation studies, we demonstrate that the CBR estimator outperforms existing methods. Lastly, the proposed method is illustrated in an analysis of AIDS clinical trial data.

个人简介:东北师范大学数学与统计学院教授、博士生导师、副院长。2006年12月博士毕业于东北师范大学,2013年12月起任东北师范大学数学与统计学院教授。2008-2010年在耶鲁大学做博士后研究,2015-2017年访问北卡大学教堂山分校。现兼任中国现场统计研究会计算统计分会副理事长,中国现场统计研究会数据科学与人工智能分会秘书长,中国概率统计学会副秘书长,吉林省现场统计研究会秘书长等。主要从事统计学的方法与应用研究,研究方向为生物统计学和生物信息学。在统计学国际顶级期刊Journal of the American Statistical Association (JASA)、医学图像著名期刊NeuroImage等发表学术论文多篇。主持并完成国家自然科学基金项目。

 

 

统计与数据科学系

2023-11-29