Machine learning in finance: practitioner’s perspective

Online interactive course

Machine Learning is increasingly applied in the finance industry, with applications ranging from credit scoring and loan approval to fraud detection and algorithmic trading. Being able to implement a machine learning (ML) algorithm is not enough to successfully apply ML in finance, since issues in financial applications of ML can be quite specific to the finance industry. Moreover, financial models are subject to different regulatory standards and scrutiny than applications of ML in, say, marketing or commerce.

In this online interactive course, we will address issues that a quant will encounter when applying machine learning models to typical problems in the finance sector.



Among topics that will be discussed are:

  • Types of ML models
  • Application areas of ML in finance
  • ML for credit models
  • Data issues in ML applications: data quantity and quality, missing values, data pre-processing, feature engineering
  • Biases in Machine Learning: how to recognise and mitigate them
  • Machine Learning for algorithmic trading: designing ML-based algo trading systems and evaluating their performance
  • Alternative data and sentiment analysis for Machine Learning applications
  • Model Risk Management and Model Validation ML models
  • Visualization and communication of ML models’ outcomes

Particular attention will be given to real finance applications / practical cases of applying ML in financial institutions: in the areas of credit risk, fraud detection and algorithmic trading.

The course is particularly suited to people who are familiar with machine learning methods (neural networks, random forest, gradient boosting etc.) but have no experience with applying these methods in financial institutions.

The course is given by Dr Svetlana Borovkova, Head of Quantitative Modelling of Probability & Partners and Professor of Quantitative Risk Management at Vrije Universiteit Amsterdam. Dr Borovkova has over 25 years of experience in quantitative finance, risk management and machine learning. She has a unique combination of academic expertise and practical experience in applying machine learning and other models in financial institutions.

One of the course sessions will be delivered by a prominent guest speaker from the finance industry.

During the course, the participants will be asked to execute a case in either credit risk or algorithmic trading, to get a working knowledge of machine learning applications in finance. The cases will be presented, discussed and extensive feedback will be given.


Registration for this course is closed.

Format: six live 2-hour Zoom sessions, combining lectures and interactive Q&A

Period: October 2020

Maximum number of participants: 25