Shortcuts | Syllabus | Notebooks
This is a graduate-level course at the Courant Institute of Mathematical Sciences, NYU. This is also a selective course in Mathematics in Finance (MFin) program.
Course information
-
Instructor: Ivailo Dimov
-
Semester: Spring 2025
-
Outline: This half-semester course (a natural sequel to the course MATH-GA. 2070 Data Science & Data-Driven Modeling) examines techniques in machine learning and computational statistics in a unified way as they are used in the financial industry. We cover supervised learning (regression and classification using linear and nonlinear models), specifically examining splines and kernel smoothers, bagging and boosting approaches; and how to evaluate and compare the performance of these machine learning models. Cross-validation and bootstrapping are important techniques from the standard machine learning toolkit, but these need to be modified when used on many financial and alternative datasets. In addition, we discuss random forests and provide an introduction to neural networks. Hands-on homework forms an integral part of the course, where we analyze real-world datasets and model them in Python using the machine learning techniques discussed in the lectures.
-
Textbook: A main good one is The Elements of Statistical Learning - Data Mining, Inference, and Prediction. See others in the syllabus.
Gradebook
Overall grade: A- (1.39/1.50)