Skip to main content

Concordia professor uses artificial intelligence to help financial institutions reduce their exposure to risk

Research by Frédéric Godin allows machine learning to find solutions for managing money smartly
May 27, 2019
|
Frédéric Godin: “My goal is to contribute to the growing body of methods and techniques in financial engineering allowing financial institutions to manage their risk.”
Frédéric Godin: “My goal is to contribute to the growing body of methods and techniques in financial engineering allowing financial institutions to manage their risk.”

How can financial institutions make the best decisions to minimize risk? While most people may not ponder this question regularly, Concordia’s Frédéric Godin argues that these choices can have a tremendous impact on our society.

Godin is assistant professor of mathematics and statistics in the Faculty of Arts and Science. He is using tools from machine learning and artificial intelligence (AI) to help investment firms and other financial institutions reduce their exposure to risk.

AI allows us to perform several tasks that are very relevant in finance

What is the main focus of your research?

Frédéric Godin: My main area of research consists of designing procedures and algorithms to allow financial institutions such as banks and insurers to measure, valuate and manage the risk they face in the best possible way.

One example consists of setting up a trading strategy aimed at purchasing financial products like stocks, financial derivatives or other assets and offsetting certain risks undertaken by the financial institution. One of the problems I consider involves optimizing the quantity and timing of such purchases using mathematical models of the behaviour of asset prices on the market.

Frédéric Godin, assistant professor of mathematics and statistics. Frédéric Godin, assistant professor of mathematics and statistics.

How does this relate to artificial intelligence?

FG: AI tools allow us to perform several tasks that are very useful and relevant in finance. These include the prediction of uncertain quantities and the automation/optimizing of decision making.

Many actors in the financial industry need to periodically make decisions, such as determining whether to buy or sell, deciding the price for an asset, selecting insurance risks admissible to underwriting or determining the amount of capital to set aside to protect the institution against a risk. These are all based on observable risk factors like interest rates, stock prices, predicted mortality of an insured, etc.

Prediction tools from machine learning and AI can be used to determine potential likely outcomes that could occur in terms of the risk factors and then learn how to make an optimal decision in that context.

What do you hope to accomplish with your work?

FG: My goal is to contribute to the growing body of methods and techniques in financial engineering allowing financial institutions to manage their risk. I want to integrate machine learning tools within these methods to improve them and make them applicable on a much larger scale.

I also aim to train highly qualified students who will be able to operate in the modern financial environment, which is rich in technology and requires advanced mathematical, statistical and computer science skills.

And how will this affect the average person?

FG: Although risk management procedures of financial institutions are not necessarily visible in the day-to-day lives of citizens, improving them strengthens the financial system and makes it more resilient to shocks. The stability of the financial system is crucial to avoid financial meltdowns like the financial crisis of 2008, which created tremendous hardship in society.

Furthermore, better risk management procedures related to financial products issued to individual investors make them less risky. Thus, financial institutions are able to offer products like segregated funds and annuities with guarantees at more competitive prices. Indeed, undergoing risk is costly due to the need to hold capital to protect the institutions against such risks.

How is Concordia active in the area of AI?

There is a lot of activity in Concordia pertaining to AI from many perspectives, including computer science, business, finance and engineering.

Faculty members from several departments are active in this area of research and tackle a diverse array of problems using these tools, such as financial risk management, software development optimization, energy demand prediction and traffic management, just to name a few.

Moreover, Montreal is one of the most active areas around the globe in machine learning and AI research, due to the presence of numerous research groups at Concordia, Université de Montréal, McGill, UQAM and the MILA research lab, as well as numerous industrial participants and stakeholders.


Find out more about
artificial intelligence research at Concordia.
 



Trending

Back to top

© Concordia University