Orchestration Framework for Financial Agents: From Algorithmic Trading to Agentic Trading

2025-12-01

AI Summary

High development barriers and domain expertise requirements hinder widespread adoption of algorithmic trading. Method: This paper proposes the first intelligent agent orchestration framework tailored for financial markets, enabling a paradigm shift from conventional algorithmic trading to autonomous multi-agent collaboration. The framework decouples strategy generation, risk management, and portfolio execution into specialized agents—Planner, Risk, and Portfolio—and integrates LLM-driven memory, audit, and backtesting agents. It supports cross-asset (equities/cryptocurrencies) and multi-horizon (hourly/minute-level) modeling. Contribution/Results: Empirical evaluation shows an annualized return of 20.42% (outperforming the S&P 500) with a Sharpe ratio of 2.63 on 2024 U.S. equity hourly data; and an 8.39% return in minute-level Bitcoin trading in 2025, significantly enhancing alpha generation. The framework advances democratization of financial AI services.

Technology Category

• Search and Optimization: Algorithm Configuration
• Multiagent Systems: Mechanism Design
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

The financial market is a mission-critical playground for AI agents due to its temporal dynamics and low signal-to-noise ratio. Building an effective algorithmic trading system may require a professional team to develop and test over the years. In this paper, we propose an orchestration framework for financial agents, which aims to democratize financial intelligence to the general public. We map each component of the traditional algorithmic trading system to agents, including planner, orchestrator, alpha agents, risk agents, portfolio agents, backtest agents, execution agents, audit agents, and memory agent. We present two in-house trading examples. For the stock trading task (hourly data from 04/2024 to 12/2024), our approach achieved a return of $20.42%$, a Sharpe ratio of 2.63, and a maximum drawdown of $-3.59%$, while the S&P 500 index yielded a return of $15.97%$. For the BTC trading task (minute data from 27/07/2025 to 13/08/2025), our approach achieved a return of $8.39%$, a Sharpe ratio of $0.38$, and a maximum drawdown of $-2.80%$, whereas the BTC price increased by $3.80%$. Our code is available on href{https://github.com/Open-Finance-Lab/AgenticTrading}{GitHub}.

Problem

• Democratizes financial intelligence through an agent orchestration framework
• Maps algorithmic trading components to specialized agents for automation
• Demonstrates improved returns and risk metrics in stock and BTC trading

Innovation

• Orchestration framework maps trading components to specialized agents
• Agents include planner, risk, portfolio, execution, and memory modules
• Democratizes financial intelligence with open-source agentic trading system