Dynamic Deep Generative Graph Models for Financial Forecasting

Borealis AI has access to a huge amount of financial data related to the stock market and is interested in leveraging recent developments in machine learning to better understand this data. Some potential questions emerging from this data are: (1) Given the closing price of a stock in the recent months, can we predict the stock returns within the next month? (2) If a stock crisis occurs, can we predict and control the spread of the crisis? (3) Given the current stock’s history, can we help reduce the risk of investment?. To answer such questions, we propose a network of related stocks based on their correlated returns. Multivariate statistical models that use tabular representations of time series can capture basic correlative structure, but we believe that innovations in machine learning known as Graph Neural Networks (GNNs) will be able to exploit this with more explicit network representation. We propose to develop a novel GNN-based algorithm on network-structured data to efficiently capture its complex structure.

Faculty Supervisor:

Graham Taylor

Student:

Elahe Ghalebi

Partner:

Borealis AI

Discipline:

Engineering

Sector:

Finance, insurance and business

University:

University of Guelph