Generative AI Enhanced Financial and Risk Analytics – Part 1

In this research project, we plan to explore potential uses of generative AI models, such as ChatGPT, to enhance financial modeling. Trusting financial advice from Generative Pre-Trained Transformer (GPT) models is a challenge due to model “hallucinations”, necessitating careful verification and validation of the output. Therefore, we plan to take an alternative approach. We would use ChatGPT to obtain a sequence of steps to solve a financial modeling problem and a list of potential data sources. Subsequently, we would validate generated output and interpret it as a sequence of computational steps to be performed with quantitative algorithms. Next, we would perform computations with quantitative finance algorithms such as Monte Carlo simulations, risk modeling and portfolio optimization. To communicate obtained computational results to a decision-maker we would feed numerical results to ChatGPT asking to interpret those. Combining new generative AI models with established numerical financial algorithms may allow us to obtain better results. By blending strengths of AI-generated problem solving methodologies with advanced quantitative finance techniques, we hope to achieve more robust and favorable investment and risk management outcomes, suggesting a hybrid approach for more effective and reliable financial decision-making in the future.

Faculty Supervisor:

Roy Kwon

Student:

Partner:

Ukrainian Catholic University

Discipline:

Engineering

Sector:

Artificial Intelligence; Finance and Insurance; Advanced Manufacturing

University:

University of Toronto

Program: