Machine Learning for Finance

URL study guide

https://studiegids.vu.nl/en/courses/2024-2025/E_FIN_MLF

Course Objective

To get acquainted with new open source techniques that can be useful for finance professionals. Techniques covered, amongst others, are:
- classification and prediction
- outlier detection
- parallelization

Course Content

It is hard to name a sector that will not be dramatically affected by machine learning or even artificial intelligence. There are many excellent courses that teach you the mechanics behind these innovations -
- helping you develop an engineering skill set. This course takes a different approach. It is aimed at people who want to deploy these tools, either in business or policy, whether through start-ups or within a large organization. While this requires some knowledge of how these tools work, that is only a small part of the equation, just as knowing how an engine works is a small part of understanding how to drive. What is really needed is an understanding of what these tools do well, and what they do badly. This course focuses on giving you a functional, rather than mechanistic, understanding. By the end, you should be an expert at identifying ideal use-cases and thereby well-placed to create new products, businesses and policies that use artificial intelligence. With the increased availability of data and cheap and fast computing power, analyses in many areas of human endeavour have become more and more data driven. Finance is no exception. Applying machine learning techniques to traditional finance questions might improve our understanding. To date, applying these techniques has been the realm of IT savvy researchers. With the increased availability of open source software these techniques are becoming widely available. To fruitfully apply them, however, finance professionals should get a much better grasp of their strengths and weaknesses and this requires first hand experience. The course goal is not to have students become experts in open source coding to apply extremely complicated machine learning models. The course should however give you a functional, rather than mechanistic, understanding. By the end, you should be an expert at identifying ideal use-cases and thereby well-placed to create new products, businesses and policies that use artificial intelligence. Moreover, your should come away from the course being able to have a meaningful discussion with experts in AI and ML fields. Topics covered so far in the undergraduate finance courses are of course many. For example, you will have seen derivative valuation models and corporate finance topics. In this course we aim to provide tools to look at these topics with a fresh angle and the aim of solving them with open source tools. We will start with tools for collaboration. Then, once we have covered the basics, we will move to the more advanced topics amongst others, supervised and unsupervised learning, and visualisation tools. We will get visits from a financial institution and see how machine learning topics are being applied in practice.

Teaching Methods

Lectures and tutorials. The course will be very interactive. Using Jupyter Notebooks we will cover the topics hands on. The aim is to getting a working knowledge of new machine learning methods so that you will be able to identify useful applications.

Method of Assessment

Assignments and a take-home exam.

Literature

We will use sources available freely on-line. Some of these sources arealso available in hard copy for ease of reference but a hard copy is notrequired for the course.

Target Audience

The course is only available to students in the Finance and Technology Honors track and the Quantitative Finance Honors track.

Entry Requirements

Python for Finance (first block) or equivalent.