STAT 210: Course Requirements

Statistics 210: Regression Analysis
Spring 2017

Course Home Page

Course Description

In this course, we learn approaches to analyzing multivariate data sets, emphasizing the analysis of variance, linear regression, and logistic regression. We learn techniques for checking the appropriateness of proposed models, such as residual analyses and case influence diagnostics, and techniques for selecting models. We emphasize data analysis over mathematical theory.

Course Objectives

To learn methods for analyzing multivariate data sets and to understand how these methods are applied in research.
To learn how to check whether proposed statistical models are appropriate for given data.
To develop proficiency in addressing complex research questions with statistical evidence.
To learn the process of data-based research by working on an independent research project.

Logistics

Lecture Times and Location: Wednesday and Friday, 1:25 - 2:40 PM, in 116 Old Chemistry Building.
Instructor: Professor Jerry Reiter, 112A Old Chemistry Building and 327 Gross Hall, 668-5227, jerry@stat.duke.edu
Instructor's Office Hours: Wednesday, 2:50 - 4:00, Thursday, 2:00 - 3:30 in Old Chem. Other hours available by appointment.
Lab Times: Monday: 11:45 AM - 1:00 PM, 116 Old Chemistry Building, and 1:25 PM - 2:40 PM in 113 Biological Sciences.
Teaching Assistants: William Eastman (office hours on Thursdays, 6:30 pm to 8:30 pm, 107 Social Sciences Building). Jianghao Li (office hours on Tuesdays, 6:00 pm to 8:00 pm, 228 Gray Building). Shinichiro Shirota (labs only).

Prerequisites

Students must have passed a 100 level STA course. Students who have taken Economics 208 are invited to speak with the instructor about differences between the courses to determine if it is worthwhile to take both.

Readings

The primary text is:

Ramsey, F. L. and Schafer, D. W. (2002), The Statistical Sleuth: Second Edition, Duxbury. Other resources are posted on the STA 210 course web site on Sakai.

Computing

We will use the statistical software package R for analyzing data. It can be downloaded for free at http://www.r-project.org/. Alternatively, R is available on the public computers on campus.

Calculator

Students don't need a calculator for this course.

Schedule of Topics

We will cover the topics in the table below. We may spend different amounts of time on each topic than shown, depending on the interests of the class participants.

Review of inference for one and two sample problems	Chapters 1, 2, 3	2 lectures
One way analysis of variance (ANOVA)	Chapters 5, 6	2 lectures
Simple linear regression model	Chapter 7, 8	2 lectures
Multiple linear regression model and general ANOVA	Chapter 9, 10, 11, 12, 13, 14	8 lectures
Time series	Chapter 15	1 lectures
Logistic regression model	Chapter 18, 19, 20, 21	3 lectures
Generalized linear models	Chapter 22	3 lectures
Missing Data	Supplemental notes	3 lectures
Algorithmic prediction methods	Supplemental notes	2 lectures

Graded work

Graded work for the course will consist of two term exams, methods and data analysis assignments, writing assignments, and a final project. There is no final exam. Students' final grades will be determined as follows:

Methods and Data Analysis Assignments
35%

Research Project
15%

Exam 1
25%

Exam 2
25%

Methods and Data Analysis Assignments	35%
Research Project	15%
Exam 1	25%
Exam 2	25%

There are no make-ups for exams, assignments, or the final project except for medical or familial emergencies or for reasons approved by the instructor before the due date. See the instructor in advance of relevant due dates to discuss possible alternatives.

Descriptions of graded work

Methods and Data Analysis Assignments:

Methods and Data Analysis assignments are posted on the Statistics 210 course web site on Sakai. Students turn in these assignments at the beginning of class on the due date. Students are permitted to work with others on the assignments, but each person must write up and turn in her or his own answers. The Methods and Data Analysis assignments include questions on the computational and the mathematical aspects of the methods that underpin the statistical models we learn during the semester, and questions that ask students to apply the modeling skills discussed during the semester.

Research Project:

For the final project, students analyze a data-based research question of their choosing, subject to the instructor's approval. The data should comprise several variables amenable to statistical analyses via regression modeling. Students can bring in their own research data sets, or they can ask the instructor for assistance with identifying appropriate data. Students present their results in a poster session. Detailed instructions are available on the Sakai site.

Term Exams:

The exams will cover mathematical and conceptual aspects of building regression models. One exam will be after the unit on multiple regression, which is approximately half way through the course. The other will be towards the end of the course.

Academic honesty

Students are expected to abide by Duke's Community Standard for all work for this course. Violations of the Standard will result in a zero grade for the relevant assignment and will be reported to the Dean of Students for adjudication. Additionally, there may be penalties to the final grade for the course. Ignorance of what constitutes academic dishonesty is not a justifiable excuse for violations.

For the exams, students are required to work alone. For the Methods and Data Analysis assignments, students may work with a study group but each student must write up and submit his or her own answers. For the final project, students are required to perform the data analysis and make the poster individually (although students can consult with the instructor and discuss ideas with other students in the class).