EUVIMEDEUVIMED entdecken
Crossref

Generative AI-Enhanced Sector-Based Investment Portfolio Construction

Oleksandr Romanko, Alina Voronina, Ruiwen Cao, Roy H. Kwon, Rafael Mendoza-Arriaga

Operations Research Forum · 15.08.2026 · 7 (3)

DOI: 10.1007/s43069-026-00686-3 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Operations Research Forum, online erschienen am 15.08.2026. Verfasst von Oleksandr Romanko und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 27 Literaturangaben. Gefördert von: Mitacs.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study investigates how large language models (LLMs) from OpenAI, Google, Anthropic, DeepSeek, and xAI can be applied to quantitative sector-based financial portfolio construction. We use LLMs to identify investable universes of stocks within S&P 500 sector indices, prompting each model to select and weight 20 stocks per sector, and evaluate the resulting portfolios against their respective indices across two out-of-sample periods: a stable market phase (January–March 2025) and a volatile phase (April–June 2025). The results reveal a strong temporal dependence in portfolio performance. Under stable market conditions, LLM-weighted portfolios frequently outperformed sector indices on both cumulative returns and risk-adjusted measures, but often underperformed during the volatile period, suggesting limited adaptability to market regime changes and high-volatility environments. When LLM-based stock selection is combined with traditional optimization techniques, portfolio outcomes improve in both performance and consistency. This multi-model, cross-provider analysis demonstrates that LLMs may complement quantitative investment processes by enhancing stock selection and interpretability, but their reliability remains sensitive to market regimes. The findings underscore the potential of hybrid AI-quantitative frameworks that integrate LLM capabilities with established optimization techniques.

Förderung

  • Mitacs

Zitieren

APA

Romanko, O., Voronina, A., Cao, R., Kwon, R. H., & Mendoza-Arriaga, R. (2026). Generative AI-Enhanced Sector-Based Investment Portfolio Construction. Operations Research Forum, 7(3). https://doi.org/10.1007/s43069-026-00686-3

Vancouver

Romanko O, Voronina A, Cao R, Kwon RH, Mendoza-Arriaga R. Generative AI-Enhanced Sector-Based Investment Portfolio Construction. Operations Research Forum. 2026;7(3). doi:10.1007/s43069-026-00686-3

BibTeX
@article{Romanko2026,
  author = {Romanko, Oleksandr and Voronina, Alina and Cao, Ruiwen and Kwon, Roy H. and Mendoza-Arriaga, Rafael},
  title = {Generative AI-Enhanced Sector-Based Investment Portfolio Construction},
  journal = {Operations Research Forum},
  year = {2026},
  volume = {7},
  number = {3},
  doi = {10.1007/s43069-026-00686-3}
}
Kostenlos starten →