onlincourse

An Introduction to Machine Learning in Quantitative Finance

Discover how machine learning can be used to solve financial data problems and create informative insights and predictions with this online course from University College London.

Duration

4 weeks

Weekly study

3 hours

Premium course

What's included?

Explore the applications of machine learning for quantitative finance

Over the past few years, machine learning (ML) has gradually permeated the financial sector, reshaping the landscape of quantitative finance as we know it.

This four-week course from University College London will demystify machine learning by uncovering its underlying mathematics and showing how to apply ML methods to real-world financial data.

Learn to use supervised learning models such as linear regression

Supervised learning is a category of machine learning that uses algorithms to classify data and create predictions.

You’ll be provided with an overview of supervised learning, as well as linear and non-linear regression with regularisation and classification. This will enable you to learn other new supervised learning algorithms in a systematic manner.

Understand how to use deep learning for predictive analytics in finance

Huge datasets are incredibly common in the financial sector, and present a significant challenge to researchers and analysts.

On this course, you’ll familiarise yourself with neural networks and understand how deep learning can be used to analyse large datasets and create accurate financial predictions. At the end of the course, you’ll put your learning into practice by tackling an empirical financial data problem using machine learning end-to-end.

Study with the experts at University College London

Your course educators are faculty members of the financial mathematics group at the UCL and Shanghai University.

With the help of their extensive research and experience, you’ll be empowered to solve real-world financial challenges through the application of modern machine learning methods.

Week 1
Introduction to the course
- Welcome to the course
  Welcome to the course! Meet our team and your fellow learners.
- What is Machine Learning?
  We give a brief introduction to Machine Learning, including its main types and examples.
- Quantitative Finance
  We will identify the challenges and promises of ML in Quantitative Finance.
- Next generation of talents in Quantitative Finance
  The impact of ML has gradually changed the landscape of quantitative finance. You will learn about the skillsets required for the next generation of quantitative researchers to adapt to the future financial world.
- End of Week 1
  After summarizing the overview of ML in quantitative finance, it is time for exercise. Let us set up the Python programming environment for your homework.
Week 2
Supervised Learning in financial applications
- Welcome to Week 2
  Learn about what tasks can be solved by supervised learning, and the main categories of supervised learning (regression v.s. classification).
- Linear Regression
  We introduce the most basic supervised learning method, i.e., linear regression.
- Regularization
  Learn about the overfitting issue and the regularization method to tackle this problem.
- Classification
  In this activity, we introduce classification, which is a supervised learning task with categorical output. You learn how to extend the regression framework to tackle classification tasks.
- End of Week 2
  We summarize a general supervised learning framework and the linear regression with regularization covered in Week 2.
Week 3
Learning derivative pricing via Deep Neural Networks
- Introduction to Week 3
  Learn about a brief introduction to deep neural networks and the underlying reason why they are powerful and universal models for data of different kinds.
- Deep Neural Networks (DNNs)
  You will learn the model architecture of artificial neural networks, including shallow neural networks and deep neural networks (DNNs).
- Training DNNs
  Introduce the back-propagation algorithm for optimizing the model parameters of Deep Neural Networks.
- Applications of DNNs to derivative pricing
  Apply DNNs to learn the derivative pricing from data using Python.
- End of Week 3
  Summarize the takeaway messages of DNNs.
Week 4
Limit order book prediction via Recurrent Neural Networks
- Introduction to Week 4
  We provide a brief introduction to Recurrent Neural Networks (RNNs), highlighting their strength in modelling time series data.
- Recurrent Neural Networks
  Find out the mathematical formulation of Recurrent Neural Network (RNN) and how to train the RNN models.
- Limit order book prediction
  We showcase how to apply RNN-based models for the tasks of limit order book prediction.
- End and next step
  We give a wrap-up of the course. This is the end activity of the course. After passing the final test, you will earn a Certificate of Achievement on this course.