Math 5813  Regression Analysis

Text Applied Linear Statistical Models   4th edition

By    Neter,Kutner,Nachtsheim, and Wasserman

Credit   3 Hours

Prrequisite  Math 3063 or equivalent, Math 4113 or equivalent 

Description  In this course, simple linear regression as well as multiple regression will be studied. Attention will be given to methods of model building, diagnostics for examining the appropriateness of a model, and remedial measures that may be helpful when the model is not appropriate. There will be extensive use of the statistical package Minitab.  An understanding of  introductory statistical concepts is needed.

 

Material Covered

 

Linear Regression with one predictor variable

Inferences in Regression Analysis

Diagnostics and Remedial Measures

Simultaneous inferences

Multiple regression models

Coefficients of partial determination

Overview of model building

Diagnostics, DFFITS, DFBETAS, and Cook’s distance

Multicollinearity

 

Learning Outcomes

 

Students will identify appropriate regression models ( L5, L7, L12)

Students will use diagnostic techniques to identify violations of assumptions of a model and use remedial measures to correct the violations. ( L11, L12)

Students will analyze residuals and their graphs to validate regression models. ( L5, L11)

Students will interpret coefficients of the models (L5, L11 )

Students will build regression models using computer-assisted selection procedures.

 

Grading

3 exams      60%

Project        20%

Final           20%

 

A project will be required of all graduate students. The project will consist of the analysis of a data set of the appendix through a collection of problems found at the end of each chapter.