R&D-Agent-Quant: A Multi-Agent Framework for Data-Centric Factors and Model Joint Optimization

2025-05-21

AI Summary

Predicting financial asset returns faces challenges of high dimensionality, non-stationarity, and strong volatility; existing quantitative pipelines suffer from insufficient automation, poor interpretability, and a fragmented separation between factor discovery and model development. This paper proposes the first data-centering multi-agent framework that establishes a closed “research–development–feedback” loop to jointly optimize factors and models. Key innovations include target-aligned prompt generation, hypothesis-driven task mapping, and a multi-armed bandit–based adaptive scheduling mechanism. The framework integrates large language model–driven code generation (Co-STEER), a live-trading backtesting engine, and a dynamic feedback evaluation architecture. Empirical results on real markets demonstrate that our approach achieves twice the annualized return of baseline factor libraries while using only 30% of the factors, significantly outperforming mainstream deep temporal models in both predictive accuracy and strategy robustness.

Technology Category

• Multiagent Systems: Mechanism Design
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Financial markets pose fundamental challenges for asset return prediction due to their high dimensionality, non-stationarity, and persistent volatility. Despite advances in large language models and multi-agent systems, current quantitative research pipelines suffer from limited automation, weak interpretability, and fragmented coordination across key components such as factor mining and model innovation. In this paper, we propose R&D-Agent for Quantitative Finance, in short RD-Agent(Q), the first data-centric multi-agent framework designed to automate the full-stack research and development of quantitative strategies via coordinated factor-model co-optimization. RD-Agent(Q) decomposes the quant process into two iterative stages: a Research stage that dynamically sets goal-aligned prompts, formulates hypotheses based on domain priors, and maps them to concrete tasks, and a Development stage that employs a code-generation agent, Co-STEER, to implement task-specific code, which is then executed in real-market backtests. The two stages are connected through a feedback stage that thoroughly evaluates experimental outcomes and informs subsequent iterations, with a multi-armed bandit scheduler for adaptive direction selection. Empirically, RD-Agent(Q) achieves up to 2X higher annualized returns than classical factor libraries using 70% fewer factors, and outperforms state-of-the-art deep time-series models on real markets. Its joint factor-model optimization delivers a strong balance between predictive accuracy and strategy robustness. Our code is available at: https://github.com/microsoft/RD-Agent.

Problem

• Automates quantitative finance strategy development via multi-agent coordination
• Addresses high dimensionality and non-stationarity in financial market prediction
• Improves factor mining and model innovation integration for better returns

Innovation

• Multi-agent framework for factor-model co-optimization
• Automates quant strategy R&D via iterative stages
• Dynamic hypothesis formulation and code-generation execution