CLEAR Seminars | Bayesian and Bayesian Structural Equation Modeling
Speaker: Professor Shing On LEUNG (University of Macau)
Chair: Professor Yuyang Cai
Time/date/Venue for Talk 1: 14:55- 16:30 / Nov 2 / SC308
Time/date/Venue for Talk 2: 10:20-12:00/ Nov 3 / SA204
Prof Shing On LEUNG got his Bachelor's degree from HKU, and Master's and Ph.D. degree from the London School of Economics. He has been working in the Faculty of Education, University of Macau, for more than 20 years. His teaching and research areas are in educational measurement and applications of statistics in education and social sciences. His recent interests are in Bayesian applications and machine learning.
Talk 1.This talk introduces basic Bayesian thinking and its differences from classical hypothesis testing. We will start from some simple examples and develop gradually into details. The differences among null hypothesis significant testing, effect sizes and Bayesian will be discussed.
Talk 2. This talk introduces Bayesian Structural Equation Modeling (BSEM). We start with different research questions in classical approach, effect size, and Bayesian analysis. Key concepts in Bayesian, such as prior, likelihood, posterior, convergence, Markov Chain Monte Carlo (MCMC), model assessment, Bayes Factors, and Bayesian Information Criteria (BIC), will be briefly mentioned before introducing BSEM. Reasons and benefits in using BSEM will be discussed.