Instructor: Merlise Clyde
| Office: | |
| Phone: | 681-8440 |
| Email: | clyde@stat.duke.edu |
| Office Hours: | Friday 2:30-3:30, or by appointment |
Teaching Assistant: Jason Duan
| Office: | |
| Phone: | 681-9390 |
| Email: | jason@stat.duke.edu |
| Office Hours: | Tuesday 4-5 (new time) |
Meeting Times: Monday & Wednesday 3:55-5:10 in 220 Social Sciences Building
This courses covers concepts of linear models from Bayesian and classical viewpoints using a vector space approach. Topics include: simple linear and multiple regression, parameter estimation and interpretation, distribution theory for ANOVA and testing, variable transformations and selection, prediction, model diagnostics, Bayesian hierarchical models, Bayes factors and model selection, and Bayesian model averaging. Extensions to Multivariate models and nonparametric mmodels if time permits. Prerequisite: Statistics 213 or equivalent. Knowledge of liinear algebra is extremely useful.
The following books will be useful for the course:
These will be at the book store or are available directly from Springer, Amazon or other sites. (Search for the best price)