STAT 220 | Last taught Spring 2024
The course focuses on the Bayesian thinking as a coherent and logical foundation, as well as a practical means, of statistical inference. The core ideas of all Bayesian methods are to design a full joint probability distribution both to describe data collection processes and to link unknown quantities of interest (e.g., some key parameters) with data, which can often incorporate complicated latent structures, and to use the Bayes theorem to derive the inferential results and make predictions. One should be sufficiently familiar with basic probability theory (STAT 110), statistical inference (STAT 111), and multivariate calculus as well as linear algebra (Math 21 a&b).