Statistical Machine Learning in Finance ECOM193

Module Content

Module Aim

The aim of this module is to give students a good understanding of the foundations of statistical machine learning and the use of selected methods in finance. There is a strong emphasis on practical application using R to model the provided datasets.

iuww520iuww520iuww520iuww520iuww520iuww520iuww520iuww520

Learning Outcomes

Upon completion of this module, students should have a good grasp of statistical machine learning and be able to undertake their own ML projects in a relatively unsupervised manner.

Module Outline

Suggested background/supplementary reading in square brackets where applicable; please refer to the Reading List given in the next section.

Introduction Course outline; what is machine learning? supervised versus unsupervised learning; the difference between classification and regression; installing and using R. [ISLR Ch.1]

Lecture 1: Machine Learning Essentials

Essential statistical knowledge; model accuracy; bias-variance tradeoff; resampling methods.

[ISLR Ch.2&5]

Lecture 2: Linear Regression

Regression basics: model fitting and diagnostics. [ISLR Ch.3]

Lecture 3: Classification

Logistic regression; discriminant analysis; naïve Bayes; nearest-neighbour classification. [ISLR Ch.4]

Online Multiple Choice Test 1 (10%)

Lecture 4: Linear Model Selection and Regularization

Automatic model selection methods; regularization and shrinkage; dimensionality reduction via principal component analysis. [ISLR Ch.6]

Midterm Coursework Assessment (20%)

Lecture 5: Towards Non-Linear Models

Polynomial regression; step functions; basis functions; regression splines; local regression;

generalized additive models. [ISLR Ch.7]

Lecture 6: Tree-Based Methods

Decision trees; bootstrap aggregation (bagging); random forests. [ISLR Ch.8]

Lecture 7: Support Vector Machines

Using support vector machines to solve classification problems. [ISLR Ch.9]