Welcome to the STAT 360 - Regression Analysis Course Website (Fall 2018)

On this website you will find:

This site will be updated as necessary throughout the semester.  To navigate the site you can use the buttons on the left.  If you have any additional items you would like to see added to the course website let me know and I will try to accommodate.

The web-based course syllabus is presented below. The Word version of the syllabus is available here.


STAT 360 - Regression Analysis - Online Syllabus

Instructor:  Dr. Brant Deppa                    
Office: 124 B Gildemeister Hall
e-mail: bdeppa@winona.edu

Office phone: 457 - 5457                                     

Office Hours:

MWF 9:00 - 12:00, 2:00 - 3:00
TTh 2:00 - 3:30

or by appointment.

Textbooks:

1) Applied Linear Regression (4th Edition) by Sanford Weisberg.

On Amazon you can rent it for $32, buy used it for around $64, or buy new for about $100. An e-book is available directly from Wiley Publishing for $113. Also see D2L for information about the 3rd edition of this text!

R Primer to accompany textbook: Computing Primer for ALR (4th ed.) by Sanford Weisberg.

2) Applied Regression Including Computing and Graphics by Dennis Cook and Sanford Weisberg

This one you can currently score used copies of for very cheap on Amazon.com (about 11 copies available when I checked) using the link above. If you cannot score a cheap used copy, don't worry - perhaps you can share with someone else in class.

Grading:

Your course grade will based on the following:

Homework and Projects 75 %
Final Exam 25 %

Homework:

Homework will consist of problems assigned from the text and others that I will write.All assignments will require the use of statistical software (JMP 13 Pro or R/R-Studio). All assignments must be submitted in Microsoft Word (.doc or .docx) format into the appropriate dropbox on D2L. You should cleanly incorporate relevant computer output into your assignments. I will demonstrate the process of cleanly capturing output from JMP and R-Studio in class and will likely put together a handout explaining the process. Working in groups of two on homework is acceptable. All course assignments must be turned in on the assigned due date and no late homework will be accepted!

Computing:

We will be using a variety of statistical software in this course.  We will primarily be using JMP 13 Pro and R/R-Studio.  R is a free statistical programming environment developed by researchers at the University of Auckland. R-Studio is an integrated development environment (IDE) for R. It includes a console, syntax-highlighting editor that supports direct code execution, as well as tools for plotting, history, debugging, and workspace management.

1) Download the most recent release of R from the Comprehensive R Archive Network (CRAN) website by clicking here

2) Download and Install R-Studio by clicking here.

3) You will also need to download a number of packages for R which are listed in the Word file: Downloading R.


4) Download and work through An Introduction to R from the Comprehensive R Archive Network under the Documentation section.


5) Update the Regression.Rdata library on your computer regularly. I will send this to you via e-mail when I have updates.

6) Download JMP 13 Pro from the campus network.

 

Tentative Course Outline and Course Notes:

The textbooks above cover nearly all of the content below, but not in the order I will be covering it. If we have time we will examine a few topics in Part IV as they set the table for DSCI 425 - Supervised Learning.

PART I - Introduction (1-2 weeks)

0 - Brief Introduction to Regression Analysis
1 - Univariate Analysis of the Response Variable (review of inference for a single numeric variable)
2 - Introduction to Regression and Conditional Distributions
3 - Introduction to Smoothing - Nonparametric Regression
4 - Bivariate Normality and Regression

PART II - Tools (8 - 10 weeks)

5 - Simple Linear Regression (SLR)
6 - SLR using Matrices
7 - Case Diagnostics for SLR: Influence and Outliers
8 - Introduction to Multiple Linear Regression (MLR)
9 - Multiple Regression - Predictors and Terms
10 - Understanding Coefficients in MLR
11 - Factors and Interactions
12 - Diagnostics I for MLR: Curvature and Nonconstant Variance
13 - Diagnostics II for MLR: Influence, Outliers, and Collinearity
14 - Response Transformations
15 - Predictor Transformations
16 - Model Selection and Prediction Accuracy
17 - Weighted Least Squares (WLS) - modeling nonconstant variance and robust regression

PART III - Introduction to Generalized Linear Models and Nonlinear Regression (2 - 3 weeks)

18 - Binomial/Logistic Regression
19 - Poisson Regression (forthcoming)
21 - Nonlinear Regression (forthcoming)

Final Exam: On your final exam you will be asked to apply all of the techiniques you have learned in the course to analyze a few real data sets.  I will give you the final during the last week of class before final exam week.  It will due sometime during finals week.