Large language models for financial and investment management: models, opportunities and challenges

Journal article : Review

The intersection of artificial intelligence (AI) and financial management has gained significant attention, particularly with the rise of large language models (LLMs). These models process vast amounts of unstructured data, offering powerful tools for financial analysis and investment decision-making. This article explores the use of LLMs in finance, focusing on recent advancements, models, and technologies, while addressing the opportunities and challenges they present. It highlights the strengths and limitations of finance-specific models in handling complex tasks and identifies key challenges such as data issues, modeling complexities, and ethical concerns, which also present opportunities for innovation. The article provides a comprehensive overview of LLMs in finance, underscoring their potential to transform the field while emphasizing the need to carefully consider their limitations and risks. The integration of LLMs into financial decision making holds significant promise, offering new possibilities for research and practical applications.

Authors: Kong, Y; Nie, Y; Dong, X; Mulvey, JM; Poor, HV

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: With Intelligence
• Journal: Journal of Portfolio Management
• Volume: 51
• Issue: 2
• Pages: 211-231
• Publication date: 2024-11-08
• Acceptance date: 2024-10-24
• DOI: 10.3905/jpm.2024.1.646
• EISSN: 2168-8656
• ISSN: 0095-4918
• Language: English
• Keywords: large language models (LLMs); financial applications; deep learning; foundation models; linguistic tasks; sentiment analysis; time series; reasoning; agent-based modeling
• Subtype: Review