ICM322-Machine Learning in Finance

Module Provider: ICMA Centre
Number of credits: 10 [5 ECTS credits]
Level:7
Terms in which taught: Spring term module
Pre-requisites:
Non-modular pre-requisites:
Co-requisites:
Modules excluded:
Current from: 2020/1

Email: m.sethi@icmacentre.ac.uk

Type of module:

Summary module description:

In this module you will learn how machine learning techniques borrowed from artificial intelligence can be used to solve common problems in finance. With the use of Python, we will explore ways in which a computer can be trained to recognise patterns in data. The focus will be on finance applications including stock price forecasting, default prediction and market sentiment analysis. 

Aims:

The module focuses on common machine learning techniques including (1) logistic regression, (2) decision trees, (3) K-nearest neighbours, (4) K-means clustering, (5) principal component analysis and (6) deep learning tools like neural networks. The emphasis will be on the use of machine learning techniques for finance applications. 

Assessable learning outcomes:

By the end of the module it is expected that students will: 

• Understand the need for a rigorous data science approach and the concepts of training data, validation data and testing data; 


• Be able to build machine learning models and interpret the models in terms of their structure and accuracy; 


• Understand how machine learning can be used to solve old and new problems in fin ance  

Additional outcomes:

The module will use the industry standard Python programming language. 

Outline content:

1. 
   

   Artificial intelligence, machine learning, deep learning 

   
   

2. 
   

   Linear and logistic regression models in Python and finance applications 

   
   

3. 
   

   Decision Tree Models in Python and finance applications 

   
   

4. 
   

   K-nearest neighbours and K-means clustering in Python and finance applications 

   
   

5. 
   

   Principal component analysis in Python and finance applications 

   
   

6. 
   

   Deep learning and neural networks in Python and finance applications 

   
   

7. 
   

   Machine learning case stu dies 

   
   

Global context:

The module covers industry standard techniques using international datasets. The concepts are applied in investment banks, central banks, hedge funds and asset management firms worldwide. 

Brief description of teaching and learning methods:

The core theory and concepts will be presented during lectures. Problem sets will be solved in workshops. 

Summative assessment- Examinations:

Summative assessment- Coursework and in-class tests:

Students will be asked to complete a report (40%) by week 2 of the summer term and one in class multiple choice test (60%) in week 7 of the spring term. 

Formative assessment methods:

Seminar questions are assigned for each class. The seminar leader will facilitate discussion and offer feedback. 

Penalties for late submission:

Penalties for late submission on this module are in accordance with the University policy. Please refer to page 5 of the Postgraduate Guide to Assessment for further information: http://www.reading.ac.uk/internal/exams/student/exa-guidePG.aspx 

Assessment requirements for a pass:

50% weighted average mark 

Reassessment arrangements:

Re assessment of individual report 

Additional Costs (specified where applicable):

Last updated: 4 April 2020