SDS7102: Linear Models and Extensions
Fall 2025
Instructor: Qiang Sun and Eric Moulines
Lectures: Tuseday, 12:00 - 13:20, CR3; Thursday, 11:00 - 12:20, CR1
Office Hours: Thursday 12:30 - 13:30, B-3.05
Email: qiang.sun@mbzuai.ac.ae & eric.moulines@mbzuai.ac.ae
TA: Ding Bai, ding.bai@mbzuai.ac.ae
Lab: Friday, 9:00 - 10:50, CR1
Please email if standard office hours times do not work for you.
Course Description
A central goal in statistics is to use data to build models that allow us to make inferences about the underlying data-generating processes or predict future observations. Although real-world problems are often complex, the linear model frequently provides a good approximation to the true data-generating process. Moreover, linear models possess elegant algebraic and geometric properties and often admit explicit formulas, offering deep insights into various aspects of modern machine learning. In our experience, the insights gained from linear models are broadly applicable, with only rare exceptions.
Textbook
- Lecture Notes
- Ding 2025, Linear Model and Extensions
- Dobson and Barnett 2018, An Introduction to Generalized Linear Models
Reference Books
- Agresti 2015, Foundations of Linear and Generalized Linear Models
- Knight 2000, Mathematical Statistics
- Hogg, McKean, Craig 2000, Introduction to Mathematical Statistics
- Evans and Rothenthal 2024, Probability and Statistics: The Science of Uncertainty
Communications
Disclaimer
All information on this class website is tentative and subject to change. Any substantive change will be accompanied with an announcement to the class via Canvas.