Applied Regression Analysis

This is an applied course in linear regression and analysis of variance (ANOVA). Topics include statistical inference in simple and multiple linear regression, residual analysis, transformations, polynomial regression, model building with real data. We will also cover one-way and two-way analysis of variance, multiple comparisons, fixed and random factors, and analysis of covariance. This is not an advanced math course, but covers a large volume of material. Requires calculus, and simple matrix algebra is helpful. We will focus on the use of, and output from, the SAS statistical software package but any statistical software can be0 used on homeworks.

STAT51200

Credit Hours:

3

Description:

This is an applied course in linear regression and analysis of variance (ANOVA). Topics include statistical inference in simple and multiple linear regression, residual analysis, transformations, polynomial regression, model building with real data. We will also cover one-way and two-way analysis of variance, multiple comparisons, fixed and random factors, and analysis of covariance. This is not an advanced math course, but covers a large volume of material. Requires calculus, and simple matrix algebra is helpful. We will focus on the use of, and output from, the SAS statistical software package but any statistical software can be0 used on homeworks.

Topics Covered:

Review of basic statistics; introduction to SAS; simple linear regression; Inference in simple linear regression; Assessing a regression model and further inference; Basic multiple regression; Full vs. Reduced model tests, polynomial regression, indicator variables; Selection and assessment of regression models; Further topics: coding data, orthogonal polynomials; One-way analysis of variance; Examination of treatment effects: contrast and Bonferroni, Scheffe, Tukey and Newman-Keuls procedures for simultaneous inference; Examining ANOVA models, transformations of the dependent variable; Random effects and introduction to two-way models; Examination of treatment effects in two-way models; analysis of covariance.

Prerequisites:

Requires calculus, and simple matrix algebra is helpful.

Applied / Theory:

80 / 20

Web Address:

www.stat.purdue.edu/~bacraig/stat512.html

Homework:

Can expect around 11 homeworks during the semester. Accepted via email at bacraig@purdue.edu.

Projects:

Required. This will be a group project that will be ongoing during the semester. It will involve the analysis of a real-world data set. A written summary and a class presentation will be due the last week of class.

Exams:

One midterm and one final exam.

Textbooks:

Official textbook information is now listed in the Schedule of Classes. NOTE: Textbook information is subject to be changed at any time at the discretion of the faculty member. If you have questions or concerns please contact the academic department.
Tentative: Required--"Applied Linear Statistical Models", Kutner, Nachstein, Neter & li, 5th Edition, ISBN 9781259064746 (also available in e-book format).
Recommended--"Applied Statistics and SAS Programming Language", Cody & Smith, 5th Edition, ISBN 0-13-146532-5.

Computer Requirements:

ProEd Minimum Computer Requirements

Other Requirements:

The class will emphasize the use of SAS statistical software. However, any statistical software that has the necessary capabilities can be used to do the homework problems.

ProEd Minimum Requirements:

view