EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Leveraging enhanced egret swarm optimization algorithm and artificial intelligence-driven prompt strategies for portfolio selection

Zhendai Huang, Zhen Zhang, Cheng Hua, Bolin Liao, Shuai Li

Scientific Reports · 04.11.2024 · 14 (1)

DOI: 10.1038/s41598-024-77925-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Scientific Reports, online erschienen am 04.11.2024. Verfasst von Zhendai Huang und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer Science and Engineering, Jishou University, Jishou, 416000, China. Die Arbeit wurde bisher 40-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In the financial field, constructing efficient investment portfolios is a focal point of research, encompassing asset selection and optimization of asset allocation. With the advancements in Large Language Models (LLMs), generative Artificial Intelligence (AI) tools have showcased capabilities never seen before. However, the black-box nature of these tools renders their outputs difficult to interpret directly, often necessitating iterative fine-tuning to align with users' expected outcomes. This study presents a structured prompt framework specifically designed for stock selection, aiming to provide direct and interpretable stock-selecting tools for investors of various levels. By creating representative scenarios and combining them into different cases for experimentation, we can explore how the construction of prompts influences the responses generated by generative AI tools. Additionally, this paper proposes a novel algorithm that combines the Nonlinear-Activated Beetle Antennae Search strategy with the Egret Swarm Optimization Algorithm (NBESOA) to address the Mean-Variance Portfolio Selection problem with Transaction Costs and Cardinality Constraints (MVPS-TCCC), utilizing real stock market data to construct portfolios based on generative AI tools recommendations. Simulation results indicate that, compared to other algorithms, NBESOA prefers optimizing portfolio configurations to achieve the highest Sharpe Ratio with the strictest constraints, bringing the outcomes closer to the portfolio's efficient frontier.

Schlagwörter

Stichwörter der Autor:innen

Bio-inspired algorithmEgret swarm optimal algorithmLarge language modelsPortfolio selectionPrompt engineering

Autor:innen und Institutionen

  • College of Computer Science and Engineering, Jishou University, Jishou, 416000, China.
  • School of Electronic Information and Electrical Engineering, Changsha University, Changsha, 410073, China.
  • College of Computer Science and Engineering, Jishou University, Jishou, 416000, China. bolinliao@jsu.edu.cn.
  • Faculty of Information Technology and Electrical Engineering, University of Oulu, 90570, Oulu, Finland. shuai.li@oulu.fi.
  • VTT-Technology Research Center of Finland, Kaitovayla 1, 90570, Oulu, Finland. shuai.li@oulu.fi.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Huang, Z., Zhang, Z., Hua, C., Liao, B., & Li, S. (2024). Leveraging enhanced egret swarm optimization algorithm and artificial intelligence-driven prompt strategies for portfolio selection. Scientific Reports, 14(1). https://doi.org/10.1038/s41598-024-77925-2

Vancouver

Huang Z, Zhang Z, Hua C, Liao B, Li S. Leveraging enhanced egret swarm optimization algorithm and artificial intelligence-driven prompt strategies for portfolio selection. Scientific Reports. 2024;14(1). doi:10.1038/s41598-024-77925-2

BibTeX
@article{Huang2024,
  author = {Huang, Zhendai and Zhang, Zhen and Hua, Cheng and Liao, Bolin and Li, Shuai},
  title = {Leveraging enhanced egret swarm optimization algorithm and artificial intelligence-driven prompt strategies for portfolio selection},
  journal = {Scientific Reports},
  year = {2024},
  volume = {14},
  number = {1},
  doi = {10.1038/s41598-024-77925-2}
}
Kostenlos starten →