Instructor: Raquel Prado, BE 365-C
Teaching Assistant: Cheng-Han Yu
Course Description
This is a graduate statistics course covering the theory and methods used to build statistical models from a Bayesian perspective. It will be assumed that students are familiar with basic ideas of Bayesian methods, including computatioons using Monte Carlo. Familiarity with a programming language (R, Matlab, Python, C, C++, F77, F95 or similar), at a level that allows the writing of relatively complex code to fit models with multiple parameters, will also be assumed. Familiarity with R is not required but very strongly recommended. Some of the topics that will be convered include: Hierarchical modeling, linear models (regression and analysis of variance), multivariate models, mixture models, predictive inference and model comparison. Prereqs: AMS-206, AMS-206B or similar course.
Classroom, Lecture Time & Office Hours
Lecture: Tu-Th 9:50-11:25am in BE 156
Office hours: Cheng-Han Yu: Mon 12-1pm, Fri 11-12pm in BE 312 C/D; Raquel Prado: Tu 11:30am-12:30pm; Wed 10:30-11:30am in BE 365C.
Textbook
Course Evaluation
There will be one midterm (TBA 40%), and two additional tests (TBA, 25%; TBA 35%). The midterm and tests will be based on the homework. They may have two parts: One to be taken in class and one to take home. The take home part will involve the analysis of a case study and/or the application of some methods taken from an article published in a leading statistical journal. You will have to turn in a a report (pdf file) following a latex or word template based on the ASA class. See ASA templates for more information.