A rigorous graduate introduction to regression analysis that develops statistical understanding alongside practical modeling skills.
Course Overview
The course covers the foundations of regression and shows how classical statistical ideas continue to inform modern machine learning and data science. Students looking for a more systematic treatment of advanced machine learning can also take Data Mining.
Assessment
- Midterm examination: 30%.
- Final examination: 40%.
- Course project: 30%.
- Homework: up to 25% bonus credit.
Both examinations begin with a two-hour closed-book written portion. A one-hour problem-solving portion follows, during which LLM tools may be used. Homework is encouraged as preparation for the examinations.
Course Project
Groups of up to three students may analyze a real-world dataset with regression and related statistical models, or pursue a more theoretical open problem. The main deliverable is a poster presentation explaining the question, methods, findings, and conclusions. Each group member should contribute and participate in the presentation.
LLM tools may be used for learning, homework, and the project. Students remain responsible for understanding their work and must disclose tool use in the project.