IE59000 - Applied Engineering Statistics: Regression
Credit Hours: 3
Instructor(s): Jack Feng
Email: fengcj@purdue.edu
Learning Objective:
Students will be able to:
- Build and interpret linear regression models
- Conduct statistical ingerence for regression parameters
- Diagnose and correct model violations
- Perform variable selection, model validation and model comparison
- Communicate findings effectively to engineering & operations audiences and leaders
Description:
This course introduces linear regression methods with an emphasis on engineering applications, model building and validation for supervised learning, and interpretation. Students learn how to develop, diagnose, and communicate regression models using real-world datasets.
Topics Covered:
- Simple and multiple linear regression
- diagnostics
- Transformations
- Multicollinearity
- Variable selection
- Predictive modeling and model validation
- Categorical predictors
Prerequisites:
- IE 23000 - Probability and Statistics in Engineering I or equivalent
- IE 33000
- MA 26500 - Linear Algebra with Applications or equivalent
Web Address:
https://purdue.brightspace.com
Homework:
13 assignments (or once/week except for mid-term and Fall Break weeks) - 39%
Project:
Term Project (team based) - 30%
Exams:
Midterm exam - 15%
Final Exam 16%
Textbooks:
Montgomery, Peck and Vining (2020) Introduction to Linear Regression Analysis, 6th Edition. Wiley