Integrating large-scale data sources, Generative AI and Large Language Models (LLMs) in advanced NLP-based techniques and quantitative models.

Picton Mahoney Asset Management (“PICTON Investments”) was founded in 2004 to provide unique investment solutions to institutional, retail, and high-net-worth investors in Canada and globally. The Quantitative Research and Risk team at PICTON Investments is dedicated to developing and maintaining models that support investment decisions and risk assessments. With the increasing complexity of financial markets and the rapid growth of both structured and unstructured data, the team faces challenges that can be addressed with more effective integration of large-scale data sources and advanced machine learning methods, including Generative AI and Large Language Models (LLMs), to strengthen insights and deliver a competitive edge in both client portfolio management and risk mitigation. This project aims to go beyond day-to-day investment operations by advancing NLP-based techniques, machine learning models, and GenAI/LLM applications to extract deeper insights from financial text and alternative data.the project requires expertise in natural language processing, generative AI, and machine learning, alongside strong foundations in statistics and finance. The primary objective involves efficiently collecting and structuring relevant financial data from a broad set of sources, including publicly available databases and sell-side analysts’ reports, in a format that supports deeper quantitative analysis. The second objective is to perform a comprehensive literature review on Natural Language Processing (NLP) and Generative AI/LLM techniques, followed by implementation of suitable methods to analyze financial text data. Key textual sources will include regulatory filings, sell-side analysts’ reports, earnings call transcripts, and market commentary. The third objective is to generate actionable investment and risk signals based on textual data and other structured datasets, followed by rigorous back-testing. To tackle these challenges, the project requires expertise in natural language processing, generative AI, and machine learning, alongside strong foundations in statistics and finance.

Faculty Supervisor:

Luis Seco;Tracy Barber

Student:

Partner:

PICTON Investments

Discipline:

Mathematics

Sector:

Finance and Insurance

University:

University of Toronto

Program: