MASS: Muli-agent simulation scaling for portfolio construction
This work addresses portfolio construction for financial investment, offering a novel method that improves adaptability and effectiveness, though it is incremental as it builds on existing LLM-based agent approaches.
The paper tackles the problem of constructing optimal portfolios by introducing the MASS framework, which uses multi-agent simulation for direct portfolio construction and demonstrates that increasing agents exponentially (up to 512) yields progressively higher excess returns, outperforming seven state-of-the-art baselines on a 2023 Chinese A-share dataset.
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 on a challenging, self-collected dataset from the 2023 Chinese A-share market show that MASS consistently outperforms seven state-of-the-art baselines. Further backtesting, stability analyses and the experiment on data leakage concerns validate its enhanced profitability and robustness. We have open-sourced our code, dataset, and training snapshots at https://github.com/gta0804/MASS/ to foster further research.