STA244 Linear Models

Course Home | Computing Links | Course Calendar | Course Info | Duke Statistics Home | Duke


Instructor: Merlise Clyde

Office:

219A Old Chemistry Building

Phone: 681-8440
Email: clyde@stat.duke.edu
Office Hours: Tuesday &Thursday10:30-11:30, or by appointment

Teaching Assistant: Laura Gunn

Office:

222 Old Chemistry Building

Phone: 684-8840
Email: laura@stat.duke.edu
Office Hours: Monday & Wednesday 2-3

Meeting Times: Tuesday & Thursday 9:10-10:25 in 219 Social Sciences Building

*Note Change in Location*


 
Texts:

TEXT

Applied Regression Including Computing and Graphics

by R. Dennis Cook and Sandford. Weisberg,  1999, John Wiley & Sons

Data sets and documentation for data sets (ASCII format)

Software:  

 

ARC

 

S-Plus 2000 / S-Plus 6
by Math Soft, Inc. ©2000

Bugslogo

Bugs and WinBugs

MRC Biostatistics Unit,Cambridge UK

 

Course Description

This courses covers concepts of linear models from Bayesian and classical viewpoints from a vector space approach. Topics include: Exploratory data analysis techniques, simple linear and multiple regression, ANOVA,variable transformations and selection, parameter estimation and interpretation, prediction, model diagnostics, Bayesian hierarchical models, Bayes factors and model selection, and Bayesian model averaging. Extensions to Multivariate models if time permits. Prerequisite: Statistics 213 or equivalent. Background in linear algebra is helpful.


Updated February 14, 2002