COMPG014 - Machine Learning with Applications in Finance

This database contains the 2017-18 versions of syllabuses. Syllabuses from the 2016-17 session are available here.

Note: Whilst every effort is made to keep the syllabus and assessment records correct, the precise details must be checked with the lecturer(s).

CodeCOMPG014
YearMSc
PrerequisitesStudents attending this module must be able to write a reasonably non-trivial computer program in MATLAB. If necessary, attend the course “Introduction to MATLAB for Finance”. In addition, the module assumes basic levels of probability theory, linear algebra, and multivariate calculus. An introduction to probability is provided by the module COMPG008.
Term2
Taught ByDaniel Fricke (100%)
AimsThis module aims at introducing students to basic ML tools, covering both supervised and unsupervised learning methods. We discuss some of the underlying principles and students will develop practical skills to use these methods in financial applications. In their coursework, students will perform their own data analyses in MATLAB.
Learning OutcomesOn successful completion of the module, students will have a good understanding of:
  1. The general background of ML methods and their differences to standard methods in Financial Econometrics and Statistics.
  2. Supervised learning methods and examples of their application in Finance.
  3. Unsupervised learning methods and examples of their application in Finance.

Content

Part 0 - Basics on Probability, Multivariate Calculus, and Linear Algebra.

To be studied independently by students.

Part 1 - General Introduction to Machine Learning.

History and background; classification of ML approaches; overview of applications.

Part 2 - Introduction and Applications of Supervised Learning.

Linear regression; model selection and regularization; feature selection; logistic regression; regression trees and forests; support vector machines; neural networks.

Part 3 - Introduction and Applications of Unsupervised Learning.

Distance/similarity measures; clustering approaches; principal component analysis.

Method of Instruction

The module is delivered via classroom-based teaching. Formative assessment in the form of coursework assignments will give students the opportunity to test their learning. 

Assessment

The course has the following assessment components:

  • Coursework (50%)
  • Written examination (2.5 hours, 50%)

To pass this course, students must:

  • Obtain an overall pass mark of 50% for all sections combined.

 

 

Resources

Reading list available via the UCL Library catalogue.