Lecture: 3 hours
Discussion: 1 hour
Simple linear regression, variable selection techniques, stepwise regression, analysis of covariance, influence measures, computing packages.
Prerequisite: STA 013 or STA 032 or STA 100
Course goals are to develop facility in the construction of multiple regression models and their application to the analysis of experimental data.
Summary of course contents:
- Standard regression model
- General linear model
- Review of matrix arithmetic
- Estimates of β and σ2
- Confidence and prediction intervals
- Decomposition of sum of squares, and lack of fit tests
- Analysis of residuals to check on validity of assumptions
- Multiple and polynomial regression-selecting correct variables
- Criteria based on R2
- Mallows's Cp
- Stepwise Regression
- Use of dummy variables
- Regression diagnostics
- Analysis of covariance
- Use of computer packages to analyze real data sets
Kutner, M.H., C.J. Nachtsheim, J. Neter and W. Li (2005). Applied Linear Statistical Models, 5th ed. McGraw-Hill, New York.
SE, QL, SL