Self-Evolving Multi-Agent Symbolic Discovery for Financial Fundamental Analysis

2026-09-26

AI Summary

This study addresses the challenges of non-stationarity, multi-perspective dynamics, and noisy feedback in applying symbolic regression to financial valuation by proposing MUFASA, a hierarchical multi-agent framework. This method decouples equation discovery from meta-coordinator reasoning to enable adaptive symbolic learning. Furthermore, it introduces a memory mechanism based on statistical performance summaries that effectively guides the search process under noisy conditions, yielding highly interpretable mathematical equations. Experimental results demonstrate that MUFASA achieves state-of-the-art performance across multinational financial datasets, significantly outperforming both conventional methods and large language models. The discovered interpretable formulas and strategy weights have been made publicly available as open-source resources.

Abstract

While symbolic regression (SR) has been successfully used in science to discover new equations, its use in financial valuation is hindered by several limitations. Whereas the natural sciences provide objectively correct relationships, financial valuation constitutes a distinct class of symbolic discovery problems, as it admits multiple valid perspectives, operates under non-stationary market conditions, and involves noisy, continuous performance signals. In this work, we propose Multi-Agent Fundamental Analysis with Symbolic Adaptive learning (MUFASA), a hierarchical multi-agent framework for symbolic discovery in finance. MUFASA introduces (1) disentangled equation discovery via specialized agents representing distinct valuation perspectives, (2) a meta-coordinator that performs hierarchical-level reasoning over market context information, and (3) a memory mechanism that reasons over statistical performance summaries (e.g., accuracy, stability, and tail risk) to guide learning under noisy feedback. Experiments across datasets from multiple countries show that MUFASA achieves state-of-the-art performance on the valuation task compared to classical finance methods, financial large language models, and SR approaches, while simultaneously producing interpretable equations, which we share with the community. We also make publicly available the distilled learnings across evolution iterations and context-dependent strategy weights, which might offer useful insights for future research on financial fundamental analysis.

Problem

• Symbolic Regression
• Financial Valuation
• Fundamental Analysis
• Non-stationary Markets
• Symbolic Discovery

Innovation

• Symbolic Regression
• Multi-Agent Framework
• Financial Fundamental Analysis
• Hierarchical Reasoning
• Memory Mechanism