An Introduction to Bayesian Analysis of Hierarchical Models

时间：2012-05-14

Title（问题）：An Introduction to Bayesian Analysis of Hierarchical Models

Speaker（报告人）：Prof. George C. Tiao, The University of Chicago、北京大学信用教授

Time（时间）：2012年5月17日（周四）下昼03:30-4:30

Place（所在）：伟易博新楼217课堂

Abstract（摘要）：Hierarchical models are widely used in statistical application, and their analyses are particularly suited for the Bayesian approach. Their increasing popularity in recent years, especially in Marketing research, is the result of phenomenal growth in computer storage and powerful computing techniques. In this report, I shall briefly go over the following topics: 1, Approaches to statistical inference: frequentist vs. Bayes; 2. MCMC computing methods; 3, Hierarchical models: fixed effect and random effects; 4, Application to atmospheric data trend analysis, and to marketing and other data; and 5, Dimension reduction, random effect models and Bayesian analysis in model building.