MASS:Multi-Agent Simulation Scaling for Portfolio Construction

Taian Guo, Haiyang SHEN, Jinsheng Huang, Junyu Luo, Zhengyang Mao, Binqi Chen, Zhuoru Chen, Luchen Liu, XIA BINGYU, XuHui Liu, Yun Ma, Ming Zhang

Keywords: Financial Market Simulations, Multi-agent systems, portfolio construction, scaling effect

Abstract: The application of LLM-based agents in financial investment has shown significant promise, yet existing approaches often require intermediate steps like predicting individual stock movements or rely on predefined, static workflows. These limitations restrict their adaptability and effectiveness in constructing optimal portfolios. In this paper, we introduce the Multi-Agent Scaling Simulation (MASS), a novel framework that leverages multi-agent simulation for direct, end-to-end portfolio construction. At its core, MASS employs a backward optimization process to dynamically learn the optimal distribution of heterogeneous agents, enabling the system to adapt to evolving market regimes. A key finding enabled by our framework is the exploration of the scaling effect for portfolio construction: we demonstrate that as the number of agents increases exponentially (up to 512), the aggregated decisions yield progressively higher excess returns. Extensive experiments are conducted on a challenging, proprietary cross-market dataset from 2023, showing that MASS gains enhanced performance over nine state-of-the-art baselines. Further backtesting, stability analyses, the experiment on data leakage concerns, and case studies validate its enhanced profitability and robustness. We have open-sourced our code, dataset, and training snapshots at https://anonymous.4open.science/r/MASS-AC96 to foster further research.

Primary Area: foundation or frontier models, including LLMs

Submission Number: 3728