STT 4830 - Linear Regression Models (3)

When Offered: Fall
An introduction to least squares estimation for both simple and multiple regression models. Linear algebra is used to represent both regression models and estimated parameters. Considerable attention is given to the analysis of variance, aptness of the model tests, residual analysis, the effects of multicollinearity, and variable selection procedures. Logistic regression, Poisson regression, and Generalized Linear Models are also introduced. Theoretical results will be explored with a modern statistical programming language.
Prerequisites: MAT 2240 and either STT 3820 or STT 3850.
[Dual-listed with STT 5830.] Dual-listed courses require senior standing; juniors may enroll with permission of the department.