STA 6207 – Regression Analysis

Instructor: Dr. Larry Winner

Office: 228 Griffin/Floyd

Phone: (352) 273-2995

Office Hours: TBA (Will be posted on webpage)

Text: Applied Regression Analysis, 2nd. Ed. by Rawlings, Pantula, Dickey

Course Description:

This course provides a survey of theory and applications in linear regression analysis. A full treatment of the linear regression model is covered, focusing on results from mathematical statistics making use of matrix algebra. Computational methods will be used to analyze datasets based on ``canned routines'' as well as a matrix language.

Tentative Topics:

· Simple Linear Regression (Chapter 1)

· Brief Introduction to Matrix Algebra (Chapter 2.1-2.8)

· Multiple Regression in Matrix Terms (Chapter 3)

· Analysis of Variance and Quadratic Forms (Chapter 4)

· Case Study (Chapter 5)

· Model Building: Selection of Independent Variables (Chapter 7)

· Polynomial Models (Chapter 8)

· Models with Class Variables (Chapter 9.6-9.7)

· Problem Areas and Diagnostics (Chapters 10,11)

· Transformations (Chapter 12)

· Intro to Nonlinear Models (Chapter 15.1-15.3)

· Logistic Regression (15.5)

· Random Coefficient Regression Models (Chapter 18.3)

Tests and Grading (Dates will be coordinated with STA 2023 & 6326):

· (Tentative) Exam 1 (7:00AM-8:25AM) - Sept. 26 – 25%

· (Tentative) Exam 2(7:00AM-8:25AM) - Oct. 31 – 25%

· (Tentative) Exam 3(7:00AM-8:25AM) - Dec. 5 – 30%

· Homework - 20%

Notes:

· Exams will be closed note. I will provide any formulas if necessary

· No make-up exams will be given. Do not plan on leaving town before Final Exam.

· Homework will be assigned on approximately a weekly basis and you will typically have 2-3 class periods to complete them. No late assignments will be accepted, and you must submit paper copies, not e-mail.

· Use e-mail sparingly. It is virtually impossible to answer technical questions via e-mail. E-mail is not a substitute for office hours/lecture.

· SAS and R code for examples in the text are available on class website.