Anas Khoukhi
Research Subjects
Anas Khoukhi has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Ilham Slimani
Mohammed First University
3 shared publications Ilham Slimani
1 shared publication
Latest Publications
- 17 November 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 356
Financial multi-agent systems: A PRISMA-based systematic review, multidimensional taxonomy, and research agenda
- 17 November 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 356
Financial multi-agent systems: A PRISMA-based systematic review, multidimensional taxonomy, and research agenda
- Open Access
- 20 September 2026
- Published by Elsevier in SSRN Electronic Journal
From Agent-Based Models to LLM-Based Financial Agents: A PRISMA-Based Systematic Review, Multidimensional Taxonomy, and Future Research Agenda
Multi-agent systems are increasingly used in finance to model market interactions, optimize sequential decisions, and coordinate specialized analytical tasks. However, the field remains fragmented across agent-based modeling, multi-agent reinforcement learning, LLM-based multi-agent systems, and hybrid architectures. This paper presents a PRISMA-based systematic review of financial multi-agent systems. Searches in Scopus and Web of Science identified 906 records. After screening, full-text assessment, and a post-extraction scope audit, 93 primary studies were retained for quantitative and thematic synthesis, while seven secondary publications were analyzed separately for contextual positioning. The results show that the literature is dominated by MARL and LLM-based multi-agent systems, followed by ABM and hybrid approaches. Portfolio management, algorithmic trading, and market simulation are the main application domains. Real financial data remain dominant, whereas textual and multimodal data are increasingly used in LLM-based systems. Evaluation is largely based on backtesting, predictive accuracy, and financial performance, with limited attention to robustness, explainability, reproducibility, human validation, regulatory alignment, and real-world deployment. The paper proposes a multidimensional taxonomy and a research agenda for transparent, robust, risk-aware, and trustworthy financial multi-agent systems.
- 20 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessFrom Agent-Based Models to LLM-Based Financial Agents: A PRISMA-Based Systematic Review, Multidimensional Taxonomy, and Future Research Agenda
Multi-agent systems are increasingly used in finance to model market interactions, optimize sequential decisions, and coordinate specialized analytical tasks. However, the field remains fragmented across agent-based modeling, multi-agent reinforcement learning, LLM-based multi-agent systems, and hybrid architectures. This paper presents a PRISMA-based systematic review of financial multi-agent systems. Searches in Scopus and Web of Science identified 906 records. After screening, full-text assessment, and a post-extraction scope audit, 93 primary studies were retained for quantitative and thematic synthesis, while seven secondary publications were analyzed separately for contextual positioning. The results show that the literature is dominated by MARL and LLM-based multi-agent systems, followed by ABM and hybrid approaches. Portfolio management, algorithmic trading, and market simulation are the main application domains. Real financial data remain dominant, whereas textual and multimodal data are increasingly used in LLM-based systems. Evaluation is largely based on backtesting, predictive accuracy, and financial performance, with limited attention to robustness, explainability, reproducibility, human validation, regulatory alignment, and real-world deployment. The paper proposes a multidimensional taxonomy and a research agenda for transparent, robust, risk-aware, and trustworthy financial multi-agent systems.
- 2 August 2026
- Published by Springer Nature
- p. 373-384
Applications of LLMs and Generative AI in Finance
- 2 August 2026
- Published by Springer Nature
- p. 373-384
Applications of LLMs and Generative AI in Finance
- 2 July 2026
- Published by Springer Nature
- p. 312-320
Toward Intelligent Financial Advisory: Hybrid Framework Leveraging LLMs, NLP, Time Series and XAI Models
- 2 July 2026
- Published by Springer Nature
- p. 312-320
Toward Intelligent Financial Advisory: Hybrid Framework Leveraging LLMs, NLP, Time Series and XAI Models