3 Hours; 3 Credits

This first course in linear models is designed to present the material related to classical regression as well as relevant modern techniques. The traditional material based on ordinary least squares is blended with the modern methods of diagnosis and combating of collinearity. In the area of selecting the optimal subset model, classical and contemporary methodologies are presented. Influence diagnostics to detect data points that exert a disproportionate influence on the regression model are also presented. In addition, procedures that are used when the assumptions of standard methodology are violated are discussed.

Prerequisite: STA 9708