统计学系系列讲座之300-301期

统计学系系列讲座之300期

时间：2018年5月28日（周一）10:00-11:00

地点：史带楼503室

主持人：夏寅 青年研究员 复旦大学管理学院统计学系

主题：Integrative Analysis of Biomedical Studies

主讲人：蔡天西 教授 哈佛大学

简介：蔡天西，哈佛大学公共卫生学院生物统计系教授，美国统计协会（American Statistical Association）会士。担任JASA, JRSSB, Life Time Data Analysis等杂志的副主编。蔡天西的研究领域包括生物标志物评估、模型选择和验证、 预测方法、个体化疾病诊断、预后和治疗、高维数据的统计推断和生存分析等。除了方法学研究之外，蔡教授在生物学、临床医学领域也做出了巨大的贡献。

摘要：Large datasets containing both a wealth of clinical and experimental data now exist as a result of the increasing adoption of electronic medical records (EMR) and availability of clinical research registries linked with specimen bio-repositories.  These datasets allow for deriving data driven classification and prediction of sub-phenotypes with high dimensional genomic and phenotypic patient level data. For example, linking the genomic and biological markers to a wide range of disease phenotypes in the EMR enables us to conduct phenome-wide association studies (PheWAS) to rigorously study genome-phenome association networks. This approach could allow the discovery ofnew subtypes of disease, along with their genetic causes. However, methodological challenges arise when performing integrative analyses of biomedical studies due to complex correlation structures, unobserved gold standard labels, and potential missingness due to study designs. I'll discuss some statistical methods that can be used to overcome such challenges. 

统计学系系列讲座之301期

时间：2018年5月29日（周二）16:00-17:00

地点：史带楼503室

主持人：张新生 教授 复旦大学管理学院统计学系

主题：A Deep Learning Perspective of Recursive Partitioning Based Methods

主讲人：Dr. Heping Zhang（张和平） Susan Dwight Bliss Professor of Biostatistics, Yale University School of Medicine

简介：Dr. Zhang published over 250 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.

摘要：Unlike any other method, deep learning has come as a forceful wave that knocked many of us to the sand, regardless of which direction we stood before. In this talk, I will begin with a brief introduction of the main steps of deep learning algorithm, and then use it to reflect on some of the statistical methods that I have developed over the past two decades, specifically classification trees for multiple binary responses (CTMBR) and multivariate adaptive splines for longitudinal data analysis (MASAL).  I will discuss the pros and cons of these methods, and offer my own perspective in our missed opportunities as well as new challenges and opportunities. 

                                                           统计学系 

2018-5-25