Financial Machine Learning · MSc-level course
Seven weeks of lecture slides, notes, and Python tutorial notebooks on machine learning for finance: regularised regression, tree ensembles, classification and credit risk, unsupervised learning, model interpretability, and neural networks — applied throughout to two running problems, market timing (time-series prediction of the equity premium) and credit-default prediction. Use the course chat to ask anything — it answers from the material itself and cites the slide, note, or tutorial it drew from.
| Week | Topic | Slides | Notes | Tutorial |
|---|---|---|---|---|
| 1 | Foundations of machine learning | Slides | Notes | View · Colab |
| 2 | Linear regression methods (Ridge, Lasso, Elastic Net) | Slides | Notes | View · Colab |
| 3 | Tree-based methods (Random Forests, Gradient Boosting) | Slides | Notes | View · Colab |
| 4 | Classification & credit risk | Slides | Notes | View · Colab |
| 5 | Unsupervised learning (PCA, clustering) | Slides | Notes | View · Colab |
| 6 | Model interpretability (SHAP, PDP) | Slides | Notes | View · Colab |
| 7 | Neural networks | Slides | Notes | View · Colab |
See also the course syllabus. The datasets are not included in the repository — see Codes/README.md for download links and setup.