统计与数据科学系系列学术报告之四百四十三-四百四十四期

时间：2024年7月16日（周二）10:00-11:00

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蒋斐宇副教授

报告人：Professor Shao Xiaofeng University of Illinois Urbana-Champaign

题目：Statistical Inference for Object-Valued Time Series

摘要：Statistical analysis of object-valued data that reside in a metric space is gradually emerging as an important branch of functional data analysis in statistics. Notable examples include networks, distributions and covariance matrices. Many object[1]valued data are collected as a time series, such as yearly age-at-death distributions for countries in Europe and daily Pearson correlation matrices for several cryptocurrencies. In this talk we introduce some recent work on statistical inference for these non-Euclidean time series. Specifically, we will cover change[1]point detection and serial independence testing. For both problems, our test statistics only depend on pairwise distance between two random objects and involve less number of tuning parameters than existing counterparts. The asymptotic theory will be presented to justify the validity of our proposed testing and estimation methods. Simulation results and real data applications are showcased to demonstrate the efficacy and versatility of our proposed procedures.

个人简介：Professor Shao Xiaofeng received his PhD degree 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, Journal of the American Statistical Association and Journal of Time Series Analysis.

统计与数据科学系系列学术报告之四百四十四期

时间：2024年7月16日（周二）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戴国榕博士

报告人：Dr. Shuang Zhou School of Mathematical and Statistical Sciences, Arizona State University, United States

题目：Bayesian Region Selection with A Spatially Dependent Global-local Shrinkage Prior

摘要：In this talk, we focus on developing a novel spatially dependent shrinkage prior for high-dimensional areal data under the Bayesian framework. The motivating problem is originated from a climate data application that aims at predicting hurricane occurrences in Atlantic basin region in United States and understanding the climate system by extracting the significant sub-regions that may be related to hurricane occurrence. A high-dimensional Bayesian Poisson model is mainly discussed in this work, with both covariate vector and coefficient vector representing certain spatial correlation patterns. Unfortunately, the current Bayesian variable selection techniques treat variables independently, which cannot capture spatially correlation structure presented in areal data. Therefore, we proposed to apply continuous shrinkage priors to Bayesian spatial models, such as the Conditional Autoregressive (CAR) model, for the purpose of region selection. Our designed prior not only achieves variable selection as the traditional methods but also retrieving the spatially correlation structure among selected signals. However, the implementation encounters new challenges due to the spatially dependence pattern, we also proposed using a stronger version of continuous shrinkage prior and several computational algorithms to obtain comparable prediction accuracy and computational efficiency. In this talk, numerical results will be presented to show a robust performance of our method for region selection under various spatial settings, and a real data application is discussed regarding the hurricane prediction for Atlantic basin region from 1950 to 2013.

个人简介：Dr. Zhou is currently an Assistant Professor in the School of Mathematical and Statistical Sciences at Arizona State University (ASU). Dr. Zhou joined ASU in August 2020, right after she obtained her PhD degree in Statistics in the Department of Statistics at Texas A&M University. Her main research interests center around developing Bayesian algorithms for large-scale data, non-standard data and spatially correlated data as well as providing theoretical guarantees. Zhou’s research also deeply focuses on interdisciplinary research in biological engineering, nuclear physics, health science, education and actuarial science, studying scientific problems with novel statistical models.