Summer School

Programme Structure

Each one-week-long module consists of ca. 15 hours of lectures, workshops, tutorials and project work. 

Students will be expected to attend on the South Kensington Campus from Monday morning onwards and each module will finish on Friday afternoon. 

The programme will be taught in English.

Entry requirements

All students are expected to have completed at least one year of undergraduate studies in quantitative fields like Mathematics, Statistics, Computing, Physics, or Engineering. Prior programming experience is not required since you will be taught the relevant skills during each module.

English requirements:

All students are required to have a good command of English, and if it is not their first language, they will need to satisfy the College requirement as follows:

• a minimum score of IELTS (Academic Test) 6.5 overall (with no less than 6.0 in any element) or equivalent.

• TOEFL (iBT) 92 overall (minimum 20 in all elements)

• CET- 4 (China) minimum score of 550

• CET- 6 (China) minimum score of 520

We ask students to bring their own laptop computer with the relevant software (R or Python - depending on the module) preinstalled. We will advise all participants prior to arrival on these requirements and will be running drop-in clinics on the first day of each module to check the preinstalled software and help with any technical issues.

Module 1: Systematic Trading

Course overview

The module “Systematics Trading” will introduce the design and deployment of systematic trading strategies used by hedge funds and other “smart money". For these, the first key ingredient are return predictions, e.g., extracted from past prices, trades, and quotes using Machine Learning methods. The second ingredient is a price impact model that quantifies the trading costs of the strategy and allows to backtest whether “paper profits” survive in the actual trading. The course will provide a hands-on introduction to both tasks.    

Learning outcomes 

For delegates who have successfully completed the course, the learning outcomes will be:  

• Design and deploy statistical models for the prediction of asset returns. 

• Implement, calibrate and leverage price impact models. 

• Combine return predictions and price impact models to build and backtest systematic trading strategies. 

Module 2: Introduction to Modern Machine Learning 

Course overview

The module “Introduction to Modern Machine Learning” will be an introductory course on modern machine learning. Students will learn basic programming skills and be introduced to key ideas from machine learning, such as linear and nonlinear methods, and how to deal with overfitting. They will also explore the concepts of supervised and unsupervised learning and the basic ideas behind deep learning.   

Learning outcomes 

For delegates who have successfully completed the course, the learning outcomes will be:  

• Understand a range of statistical and mathematical techniques to manipulate empirical data sets. 

• Implement machine learning algorithms. 

• Apply learnt techniques to real life data sets. 

Module 3: Machine Learning and Statistics for Time Series Analysis

Course overview

In the module “Machine Learning and Statistics for Time Series Analysis”, students will embark on a comprehensive exploration of probability theory and time series analysis combining ideas from modern statistics and recent advances in machine learning. They will learn how to handle trends and seasonality within time series data and develop expertise in forecasting time series models. The theoretical developments will be applied to various data sets with a particular focus on financial data. Students will also learn how risk measures such as value-at-risk and expected shortfall can be computed.

Learning outcomes 

For delegates who have successfully completed the course, the learning outcomes will be:  

• Understand a range of statistical and mathematical techniques to manipulate empirical data sets. 

• Implement machine learning algorithms. 

• Explain time series modelling. 

• Apply learnt techniques to real life data sets. 

Apply now!