{"title":"MarS:由生成式基础模型支持的金融市场模拟引擎","authors":"Junjie Li, Yang Liu, Weiqing Liu, Shikai Fang, Lewen Wang, Chang Xu, Jiang Bian","doi":"arxiv-2409.07486","DOIUrl":null,"url":null,"abstract":"Generative models aim to simulate realistic effects of various actions across\ndifferent contexts, from text generation to visual effects. Despite efforts to\nbuild real-world simulators, leveraging generative models for virtual worlds,\nlike financial markets, remains underexplored. In financial markets, generative\nmodels can simulate market effects of various behaviors, enabling interaction\nwith market scenes and players, and training strategies without financial risk.\nThis simulation relies on the finest structured data in financial market like\norders thus building the finest realistic simulation. We propose Large Market\nModel (LMM), an order-level generative foundation model, for financial market\nsimulation, akin to language modeling in the digital world. Our financial\nMarket Simulation engine (MarS), powered by LMM, addresses the need for\nrealistic, interactive and controllable order generation. Key objectives of\nthis paper include evaluating LMM's scaling law in financial markets, assessing\nMarS's realism, balancing controlled generation with market impact, and\ndemonstrating MarS's potential applications. We showcase MarS as a forecast\ntool, detection system, analysis platform, and agent training environment. Our\ncontributions include pioneering a generative model for financial markets,\ndesigning MarS to meet domain-specific needs, and demonstrating MarS-based\napplications' industry potential.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"22 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"MarS: a Financial Market Simulation Engine Powered by Generative Foundation Model\",\"authors\":\"Junjie Li, Yang Liu, Weiqing Liu, Shikai Fang, Lewen Wang, Chang Xu, Jiang Bian\",\"doi\":\"arxiv-2409.07486\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Generative models aim to simulate realistic effects of various actions across\\ndifferent contexts, from text generation to visual effects. Despite efforts to\\nbuild real-world simulators, leveraging generative models for virtual worlds,\\nlike financial markets, remains underexplored. In financial markets, generative\\nmodels can simulate market effects of various behaviors, enabling interaction\\nwith market scenes and players, and training strategies without financial risk.\\nThis simulation relies on the finest structured data in financial market like\\norders thus building the finest realistic simulation. We propose Large Market\\nModel (LMM), an order-level generative foundation model, for financial market\\nsimulation, akin to language modeling in the digital world. Our financial\\nMarket Simulation engine (MarS), powered by LMM, addresses the need for\\nrealistic, interactive and controllable order generation. Key objectives of\\nthis paper include evaluating LMM's scaling law in financial markets, assessing\\nMarS's realism, balancing controlled generation with market impact, and\\ndemonstrating MarS's potential applications. We showcase MarS as a forecast\\ntool, detection system, analysis platform, and agent training environment. Our\\ncontributions include pioneering a generative model for financial markets,\\ndesigning MarS to meet domain-specific needs, and demonstrating MarS-based\\napplications' industry potential.\",\"PeriodicalId\":501478,\"journal\":{\"name\":\"arXiv - QuantFin - Trading and Market Microstructure\",\"volume\":\"22 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-04\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - QuantFin - Trading and Market Microstructure\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07486\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - QuantFin - Trading and Market Microstructure","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07486","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
生成模型旨在模拟从文本生成到视觉效果等不同情境下各种行为的真实效果。尽管人们一直在努力构建真实世界的模拟器,但在金融市场等虚拟世界中利用生成模型的探索仍然不足。在金融市场中,生成模型可以模拟各种行为的市场效应,实现与市场场景和玩家的互动,并在没有金融风险的情况下训练策略。我们提出了大型市场模型(Large MarketModel,LMM),这是一种订单级生成基础模型,用于金融市场模拟,类似于数字世界中的语言建模。我们的金融市场仿真引擎(MarS)以 LMM 为动力,满足了对逼真、互动和可控订单生成的需求。本文的主要目标包括评估 LMM 在金融市场中的缩放规律、评估 MarS 的真实性、平衡可控生成与市场影响,以及展示 MarS 的潜在应用。我们展示了作为预测工具、检测系统、分析平台和代理培训环境的 MarS。我们的贡献包括开创金融市场生成模型、设计 MarS 以满足特定领域的需求,以及展示基于 MarS 的应用的行业潜力。
MarS: a Financial Market Simulation Engine Powered by Generative Foundation Model
Generative models aim to simulate realistic effects of various actions across
different contexts, from text generation to visual effects. Despite efforts to
build real-world simulators, leveraging generative models for virtual worlds,
like financial markets, remains underexplored. In financial markets, generative
models can simulate market effects of various behaviors, enabling interaction
with market scenes and players, and training strategies without financial risk.
This simulation relies on the finest structured data in financial market like
orders thus building the finest realistic simulation. We propose Large Market
Model (LMM), an order-level generative foundation model, for financial market
simulation, akin to language modeling in the digital world. Our financial
Market Simulation engine (MarS), powered by LMM, addresses the need for
realistic, interactive and controllable order generation. Key objectives of
this paper include evaluating LMM's scaling law in financial markets, assessing
MarS's realism, balancing controlled generation with market impact, and
demonstrating MarS's potential applications. We showcase MarS as a forecast
tool, detection system, analysis platform, and agent training environment. Our
contributions include pioneering a generative model for financial markets,
designing MarS to meet domain-specific needs, and demonstrating MarS-based
applications' industry potential.