COMM 486I - Applications of AI in Finance

This course examines the applications of artificial intelligence (AI) and machine learning (ML) in modern finance, focusing on their role in investment strategies, market analysis, and financial decision-making. Students will explore AI-driven approaches to alternative data processing, factor investing, technical analysis, and fundamental analysis. Through case studies, academic research, and hands-on implementation, the course demonstrates how AI models enhance financial forecasting, trading strategies, and portfolio management.

Students will apply key AI tools, including machine learning for asset pricing, convolutional neural networks (CNNs) for trading signal extraction, natural language processing (NLP) for alternative data analysis, and decision trees, random forests, and recurrent neural networks (RNNs) for stock valuation. Practical applications include smart beta strategies, AI-powered trading, sentiment analysis of financial transcripts, and unstructured data-driven fundamental analysis. All applications will be using real-world data from financial markets and simulate realistic decision-making scenarios from leading financial institutions.

This is a highly applied course, integrating Python programming, Microsoft Copilot for Excel, and other AI tools used in finance. Students are expected to have a strong foundation in quantitative methods and prior experience with Python programming, as group projects will require coding proficiency for data analysis and model implementation. A solid understanding of linear algebra, differential calculus, and statistical analysis will help students grasp the mathematical underpinnings of AI models. A strong analytical mindset and a willingness to tackle complex, data-driven financial problems are essential for success in this course.

Learning objectives

By the end of this course, students will be able to:

1. Determine the potential of artificial intelligence (AI) and machine learning (ML) in transforming financial services and markets.
2. Design AI-driven solutions for key financial tasks, including portfolio optimization, investment analysis, and financial forecasting.
3. Create natural language processing (NLP) applications to analyze alternative financial data sources such as earnings calls, news sentiment, and textual reports.
4. Generate Python programs for financial data analysis, emphasizing machine learning model implementation and AI-driven insights.
5. Build computational codes that replicates academic research papers applying AI and ML concepts in practical financial scenarios.
6. Judge the usability of Microsoft Copilot for Excel in automating financial analysis, improving workflows, and enhancing decision-making.
7. Compare generative AI applications in finance, including industry-specific models like BloombergGPT and their impact on research and decision-making.

Admissions

Admission to COMM 486I is by application only and is restricted to students entering Year 4 in any undergraduate program at UBC. To apply, please email Jan Bena at jan.bena@sauder.ubc.ca to express your interest, and you will be directed to the course application page.

For the 2025 Winter Session, the deadline to apply is Friday, May 16, 2025 at 4:00pm. Successful applicants will be notified of acceptance by mid-June.

The course requires a combination of knowledge in three key areas: (1) finance fundamentals, (2) computer programming, and (3) quantitative skills in statistics or econometrics. This knowledge can be acquired through relevant coursework—for example:

• Finance: COMM 298, COMR 473, COMM/COEC 370, COMM/COEC 371, COMM 374, or equivalent
• Computer Programming: CPSC 103, CPSC 110, CPSC 310, COMM 337, COMM 475, or equivalent
• Quantitative Skills: COMM 414, COMM 415, ECON 328, or equivalent

As part of the admission process, applicants will be asked to provide information about their coursework or experience in these three areas.

Course credits: 3