Course Agenda

Agenda

Machine Learning in Finance - London

Course Agenda

Day 1 

08:30

Registration and refreshments

09:00

Machine Learning: Opportunities, Limitations and Applications

  • Introduction to machine learning in finance 

  • Neural networks 

  • Opportunities and limitations 

  • Regulation 

  • Use cases  

10:30

Morning break

11:00

Machine Learning Models 

  • Supervised learning

  • Unsupervised learning 

  • Reinforcement learning 

  • Deep learning 

  • Advanced machine learning models 

12:30

Lunch

13:30

Machine Learning for Risk Management

  • Machine learning in banking, risk management & modelling

  • Analysis of rare events

             - Labeled, but unbalanced data: the case of credit card fraud
             - Unlabeled data: operational risk events
             - Discriminative vs. generative models: when are distributions relevant? 

  • Analysing networks of payments

  • Time series spikes and breakouts

15:00

Afternoon break

15:30

Machine Learning in Finance: Putting it into Practice 

  • ML for quantitative investment: challenges and opportunities 

  • Where does ML have more chance of winning? 

  • Start with a baseline model 

  • Two examples: 

    • Forecasting dividend ends 

    • Building a stock selection model 

17:00

End of Day 1

Day 2

08:30

Refreshments

09:00

Machine Learning in Finance: Future Opportunities 

  • Personalised Banking 

  • Credit Risk 

  • NLP 

  • Reinforcement learning for trading 

  • Operational Optimisation 

  • Supply Chain Finance

  • Interpretation of neural networks 

10:00

Morning break

10:30

Unsupervised Learning Applied to Trade Data 

  • Dealing with trade data 

  • Autoencoders 

  • Unsupervised learning 

  • Implementation insights

12:00

Lunch

13:00

Automated ML for Time Series Predictions and Applications in Finance 

  • What is AutoML and how can it be applied to time series data? 

  • How to use full pipeline optimization (including cleaning/filtering/feature engineering/selection and model selection) to produce optimal forecasts 

  • How to use genetic programming techniques to automatically generate signals 

  • How to evaluate quality signals and scan alternative data for alpha

14:30

Afternoon break

15:00

Alternative Data for Investors 

  • Define what data is, discussing the various challenges associate with its usage 

  • Introduce the topic of natural language processing (NLP), and how it can be used to structure text based data and various Python tools which can be used to accomplish this 

  • Discuss a number of different investor use cases for alternative data for a number of different types of datasets, including those based on images and text 

  • Use cases include forecasting EPS for publicly traded European retailers from satellite imagery, using FX flow data to trade FX spot and also machine readable news to trade FX 

  • The content is based on parts of The Book of Alternative Data, by Alex Denev and Saeed Amen, which will be published on Wiley in 2020 

16:30

End of course

 

Machine Learning in Finance - London, 4 - 5 September

View agenda