Gate研究院:多智能体LLM交易框架在BTC回测中显著跑赢Buy & Hold策略

Odaily星球日报讯 Gate 研究院近日发布的《基于 Multi-Agent LLM 的 BTC 交易框架研究与回测分析》指出,相比单一 LLM 直接生成交易信号,多智能体(Multi-Agent)LLM 架构更接近真实金融机构的投研流程,能够通过分析师、研究员、交易员与风控团队之间的协作与辩论,提升交易决策的透明度与风险控制能力。研究基于 TradingAgents 框架,对 BTC 市场构建了适用于加密场景的 AI 交易系统,并引入技术分析、新闻分析、情绪分析与宏观/链上分析等多类智能体角色。
研究基于 BTC/USDT 1 小时级别数据,对 TradingAgents-BTC 策略进行了历史回测。结果显示,该策略在测试区间内取得 +20.25% 总收益,显著优于同期 Buy & Hold 的 -7.89%,同时最大回撤控制在 -17.41%,低于 Buy & Hold 的 -27.06%。研究认为,多智能体框架在震荡与下跌阶段能够通过 Sell/Underweight 与 Flat 状态减少部分风险暴露,并在市场反弹阶段重新切入多头,从而提升整体风险收益表现。
报告指出,Multi-Agent LLM 框架在加密交易场景中展现出一定应用潜力,但当前回测周期仅覆盖约三个月,且 1 小时级别交易仍可能受到手续费、滑点与信号延迟影响。未来仍需在更长历史周期、不同市场环境与更多资产类别中进一步验证策略稳定性与泛化能力。
Disclaimer: OKX Orbit content is provided for informational purposes only. Learn more
Replies
Related Flash News
Bitcoin's volatility is near its lowest point since 2023, and the crypto market is entering a 'calm breather'
Binance将下架APT/ETH、CTSI/BTC等现货交易对
BTC OG内幕巨鲸:ZEC空单由亏转盈
Italian energy company Alps Blockchain has converted an idle power plant in Bolivia into a Bitcoin mining farm
The ETH/BTC exchange rate continues to decline, currently quoted at 0.02713
10x Research: Most Bitcoin treasury companies have significantly compressed their NAV premiums and may enter an era of negative premiums
Bitcoin has entered a high-risk range, and the continued withdrawal of institutional funds highlights concerns about selling pressure
Data: Whales opened $40.26 million in BTC short positions and held $33.3 million in ZEC long positions
Bitunix Analyst: Hormuz is 'fighting while talking,' and the market is beginning to realize that real uncertainty has never disappeared
The survey shows that 16% of Brazilian investors have allocated cryptocurrencies, while 56% of non-investors intend to enter the market


