一、时间安排

二、课程简介

Model Averaging

Xinyu Zhang

Chinese Academy of Sciences

 The past two decades witnessed significant growth of the literature on model averaging from the frequentist perspective. Some important progresses have been made. In this short course, I will introduce some frequentist model averaging approaches, which include model averaging based on information criteria, least squares model averaging, and adaptive model averaging, among others. The large sample properties such as asymptotic optimality and asymptotic distribution will be focused on. Several recent topics and future research directions on model averaging will also be discussed.

Statistical Analysis of Bioinformatics Data

Hongyu Miao

UTHealth School of Public Health

Bioinformatics has been playing an important role in biomedical research and clinical practice within the past decade. Despite the development and application of numerous methods and tools for a variety of biomedical problems, a significant amount of efforts have been dedicated to molecular-level information extraction and analysis (e.g., genome and proteome) in the bioinformatics field. This short course aims to provide an introduction to basic problems and methods in bioinformatics research as well as to cover several selected cutting-edge methods and topics. Particular examples on single-cell data normalization, differentially expressed gene identification, pathway analysis, and network analysis will be given, with necessary illustration of the related computing techniques in R/Bioconductor. Participants of this course are expected to gain experiences with both methodology and computing technique development.

Statistical Inference for Big Data

Haiying Wang

University of Connecticut

Extraordinary amounts of data are being produced in many branches of science. While providing numerous opportunities for researchers to tackle more complicated research questions, the rapidly growing data volume undoubtedly poses various new challenges. One common challenge from the statistical perspective is how to obtain useful information from massive data with limited computing facilities. Since the advance of computing technologies lags far behind the exponential growth of database sizes, it is crucial to draw useful conclusions from massive data using available fixed computing power. The course introduces the recently developed techniques for this purpose, with an emphasis on subsampling method. It will give students an overview of the emerging field and point out new research opportunities.