报告人：宋心远(香港中文大学)

邀请人：潘灯

报告主题：Bayesian Models for Complex Data Analysis

报告摘要：This series of talks covers recent developments in Bayesian modeling and data analysis. It includes four talks. Talk 1 introduces principles of Bayesian inference, including Bayesian updating, prior specification, posterior derivation, data augmentation, MCMC algorithm, and sensitivity analysis. Apart from estimation, Bayesian model/variable selection methods, such as Bayes factor, deviance information criterion, and Bayesian Lasso, are described. Talk 2 introduces extended Generalized Linear Models for ordinal and nominal data. Talk 3 introduces extended Generalized Linear Mixed Effect Models for discrete data. Talk 4 introduces Bayesian models for heterogenous, hierarchical and missing data. Real applications are used to illustrate the methodologies.

报告人简介：宋心远，香港中文大学统计系主任，教授。研究领域是潜变量模型，非参数和半参数模型，贝叶斯方法，统计计算和生存分析。宋心远教授是统计学和心理学多个国际期刊的副主编，包括Psychometrika, Structural Equation Modeling, Biometrics, Canadian Journal of Statistics, Computational Statistics and Data Analysis等。宋心远教授已在统计学，生物统计学，心理学和其他交叉学科领域的国际期刊上发表超过140篇论文。

报告题目（一）：Principle of Bayesian Inference

报告时间：2021年1月27日（星期三）14:00-16:00

报告地点：腾讯会议 987 2161 0473

报告题目（二）：Extended Generalized Linear Models for Ordinal and Nominal Data

报告时间：2021年1月28日（星期四）14:00-16:00

报告地点：腾讯会议 987 2161 0473

报告题目（三）：Extended Generalized Linear Mixed Effect Models for Discrete Data

报告时间：2021年1月29日（星期五）14:00-16:00

报告地点：腾讯会议 987 2161 0473

报告题目（四）：Bayesian Models for Heterogenous, Hierarchical and Missing data

报告时间：2021年1月30日（星期六）14:00-16:00

报告地点：腾讯会议 987 2161 0473