GENERATIVE PRETRAINED TRANSFORMERS FOR INVESTOR-CENTRIC PORTFOLIO CONSTRUCTION

Dimitrios Papakyriakopoulos () and Manolis Kritikos ()
Additional contact information
Dimitrios Papakyriakopoulos: Athens University of Economics and Business, Greece
Manolis Kritikos: Athens University of Economics and Business, Greece

Journal of Information Systems & Operations Management, 2025, vol. 19, issue 1, 226-240

Abstract: The objective of this paper is to examine the application of generative artificial intelligence in personalized portfolio construction and evaluate its performance relative to traditional benchmarks. A generative AI model, specifically OpenAI’s GPT-4o, was employed to construct investment portfolios for ten virtual investor profiles over a fixed three-month investment horizon. The methodology involved prompting the model to create portfolio allocations, followed by performance evaluation using financial metrics including total return, volatility, beta, Sharpe ratio, and maximum drawdown. All AI-generated portfolios outperformed the S&P 500 index over the investment period, demonstrating stronger risk- adjusted returns and lower drawdowns. These results highlight the potential of large language models to synthesize financial data and produce competitive investment strategies. The study contributes to the growing body of research on AI-driven decision- making in finance and provides a foundation for the development of generative models tailored to asset and wealth management

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.rebe.rau.ro/RePEc/rau/jisomg/SU25/JISOM-SU25-A15.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:rau:jisomg:v:19:y:2025:i:1:p:226-240

Access Statistics for this article

More articles in Journal of Information Systems & Operations Management from Romanian-American University Contact information at EDIRC.
Bibliographic data for series maintained by Alex Tabusca ().