This paper develops a mathematical framework for building a position in a stock over a fixed period of time while in competition with one or more other traders doing the same thing. We develop a game-theoretic framework that takes place in the space of trading strategies where action sets are trading strategies and traders try to devise best-response strategies to their adversaries. In this setup trading is guided by a desire to minimize the total cost of trading arising from a mixture of temporary and permanent market impact caused by the aggregate level of trading including the trader and the competition. We describe a notion of equilibrium strategies, show that they exist and provide closed-form solutions.
{"title":"Optimal position-building strategies in Competition","authors":"Neil A. Chriss","doi":"arxiv-2409.03586","DOIUrl":"https://doi.org/arxiv-2409.03586","url":null,"abstract":"This paper develops a mathematical framework for building a position in a\u0000stock over a fixed period of time while in competition with one or more other\u0000traders doing the same thing. We develop a game-theoretic framework that takes\u0000place in the space of trading strategies where action sets are trading\u0000strategies and traders try to devise best-response strategies to their\u0000adversaries. In this setup trading is guided by a desire to minimize the total\u0000cost of trading arising from a mixture of temporary and permanent market impact\u0000caused by the aggregate level of trading including the trader and the\u0000competition. We describe a notion of equilibrium strategies, show that they\u0000exist and provide closed-form solutions.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
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.
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。
{"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":"https://doi.org/arxiv-2409.07486","url":null,"abstract":"Generative models aim to simulate realistic effects of various actions across\u0000different contexts, from text generation to visual effects. Despite efforts to\u0000build real-world simulators, leveraging generative models for virtual worlds,\u0000like financial markets, remains underexplored. In financial markets, generative\u0000models can simulate market effects of various behaviors, enabling interaction\u0000with market scenes and players, and training strategies without financial risk.\u0000This simulation relies on the finest structured data in financial market like\u0000orders thus building the finest realistic simulation. We propose Large Market\u0000Model (LMM), an order-level generative foundation model, for financial market\u0000simulation, akin to language modeling in the digital world. Our financial\u0000Market Simulation engine (MarS), powered by LMM, addresses the need for\u0000realistic, interactive and controllable order generation. Key objectives of\u0000this paper include evaluating LMM's scaling law in financial markets, assessing\u0000MarS's realism, balancing controlled generation with market impact, and\u0000demonstrating MarS's potential applications. We showcase MarS as a forecast\u0000tool, detection system, analysis platform, and agent training environment. Our\u0000contributions include pioneering a generative model for financial markets,\u0000designing MarS to meet domain-specific needs, and demonstrating MarS-based\u0000applications' industry potential.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jialun Cao, David Šiška, Lukasz Szpruch, Tanut Treetanthiploet
We analyse the regret arising from learning the price sensitivity parameter $kappa$ of liquidity takers in the ergodic version of the Avellaneda-Stoikov market making model. We show that a learning algorithm based on a regularised maximum-likelihood estimator for the parameter achieves the regret upper bound of order $ln^2 T$ in expectation. To obtain the result we need two key ingredients. The first are tight upper bounds on the derivative of the ergodic constant in the Hamilton-Jacobi-Bellman (HJB) equation with respect to $kappa$. The second is the learning rate of the maximum-likelihood estimator which is obtained from concentration inequalities for Bernoulli signals. Numerical experiment confirms the convergence and the robustness of the proposed algorithm.
{"title":"Logarithmic regret in the ergodic Avellaneda-Stoikov market making model","authors":"Jialun Cao, David Šiška, Lukasz Szpruch, Tanut Treetanthiploet","doi":"arxiv-2409.02025","DOIUrl":"https://doi.org/arxiv-2409.02025","url":null,"abstract":"We analyse the regret arising from learning the price sensitivity parameter\u0000$kappa$ of liquidity takers in the ergodic version of the Avellaneda-Stoikov\u0000market making model. We show that a learning algorithm based on a regularised\u0000maximum-likelihood estimator for the parameter achieves the regret upper bound\u0000of order $ln^2 T$ in expectation. To obtain the result we need two key\u0000ingredients. The first are tight upper bounds on the derivative of the ergodic\u0000constant in the Hamilton-Jacobi-Bellman (HJB) equation with respect to\u0000$kappa$. The second is the learning rate of the maximum-likelihood estimator\u0000which is obtained from concentration inequalities for Bernoulli signals.\u0000Numerical experiment confirms the convergence and the robustness of the\u0000proposed algorithm.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194498","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Financial time series often exhibit low signal-to-noise ratio, posing significant challenges for accurate data interpretation and prediction and ultimately decision making. Generative models have gained attention as powerful tools for simulating and predicting intricate data patterns, with the diffusion model emerging as a particularly effective method. This paper introduces a novel approach utilizing the diffusion model as a denoiser for financial time series in order to improve data predictability and trading performance. By leveraging the forward and reverse processes of the conditional diffusion model to add and remove noise progressively, we reconstruct original data from noisy inputs. Our extensive experiments demonstrate that diffusion model-based denoised time series significantly enhance the performance on downstream future return classification tasks. Moreover, trading signals derived from the denoised data yield more profitable trades with fewer transactions, thereby minimizing transaction costs and increasing overall trading efficiency. Finally, we show that by using classifiers trained on denoised time series, we can recognize the noising state of the market and obtain excess return.
{"title":"A Financial Time Series Denoiser Based on Diffusion Model","authors":"Zhuohan Wang, Carmine Ventre","doi":"arxiv-2409.02138","DOIUrl":"https://doi.org/arxiv-2409.02138","url":null,"abstract":"Financial time series often exhibit low signal-to-noise ratio, posing\u0000significant challenges for accurate data interpretation and prediction and\u0000ultimately decision making. Generative models have gained attention as powerful\u0000tools for simulating and predicting intricate data patterns, with the diffusion\u0000model emerging as a particularly effective method. This paper introduces a\u0000novel approach utilizing the diffusion model as a denoiser for financial time\u0000series in order to improve data predictability and trading performance. By\u0000leveraging the forward and reverse processes of the conditional diffusion model\u0000to add and remove noise progressively, we reconstruct original data from noisy\u0000inputs. Our extensive experiments demonstrate that diffusion model-based\u0000denoised time series significantly enhance the performance on downstream future\u0000return classification tasks. Moreover, trading signals derived from the\u0000denoised data yield more profitable trades with fewer transactions, thereby\u0000minimizing transaction costs and increasing overall trading efficiency.\u0000Finally, we show that by using classifiers trained on denoised time series, we\u0000can recognize the noising state of the market and obtain excess return.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We propose that a tree-like hierarchical structure represents a simple and effective way to model the emergent behaviour of financial markets, especially markets where there exists a pronounced intersection between social media influences and investor behaviour. To explore this hypothesis, we introduce an agent-based model of financial markets, where trading agents are embedded in a hierarchical network of communities, and communities influence the strategies and opinions of traders. Empirical analysis of the model shows that its behaviour conforms to several stylized facts observed in real financial markets; and the model is able to realistically simulate the effects that social media-driven phenomena, such as echo chambers and pump-and-dump schemes, have on financial markets.
{"title":"Simulation of Social Media-Driven Bubble Formation in Financial Markets using an Agent-Based Model with Hierarchical Influence Network","authors":"Gonzalo Bohorquez, John Cartlidge","doi":"arxiv-2409.00742","DOIUrl":"https://doi.org/arxiv-2409.00742","url":null,"abstract":"We propose that a tree-like hierarchical structure represents a simple and\u0000effective way to model the emergent behaviour of financial markets, especially\u0000markets where there exists a pronounced intersection between social media\u0000influences and investor behaviour. To explore this hypothesis, we introduce an\u0000agent-based model of financial markets, where trading agents are embedded in a\u0000hierarchical network of communities, and communities influence the strategies\u0000and opinions of traders. Empirical analysis of the model shows that its\u0000behaviour conforms to several stylized facts observed in real financial\u0000markets; and the model is able to realistically simulate the effects that\u0000social media-driven phenomena, such as echo chambers and pump-and-dump schemes,\u0000have on financial markets.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The year 2024 witnessed a major development in the cryptocurrency industry with the long-awaited approval of spot Bitcoin exchange-traded funds (ETFs). This innovation provides investors with a new, regulated path to gain exposure to Bitcoin through a familiar investment vehicle (Kumar et al., 2024). However, unlike traditional ETFs that directly hold underlying assets, Bitcoin ETFs rely on a creation and redemption process managed by authorized participants (APs). This unique structure introduces distinct characteristics in terms of premium/discount behavior compared to traditional ETFs. This paper investigates the premium and discount patterns observed in Bitcoin ETFs during first four-month period (January 11th, 2024, to May 17th, 2024). Our analysis reveals that these patterns differ significantly from those observed in traditional index ETFs, potentially exposing investors to additional risk factors. By identifying and analyzing these risk factors associated with Bitcoin ETF premiums/discounts, this paper aims to achieve two key objectives: Enhance market understanding: Equip and market and investors with a deeper comprehension of the unique liquidity risks inherent in Bitcoin ETFs. Provide a clearer risk management frameworks: Offer a clearer perspective on the risk-return profile of digital asset ETFs, specifically focusing on Bitcoin ETFs. Through a thorough analysis of premium/discount behavior and the underlying factors contributing to it, this paper strives to contribute valuable insights for investors navigating the evolving landscape of digital asset investments
{"title":"Bitcoin ETF: Opportunities and risk","authors":"Di Wu","doi":"arxiv-2409.00270","DOIUrl":"https://doi.org/arxiv-2409.00270","url":null,"abstract":"The year 2024 witnessed a major development in the cryptocurrency industry\u0000with the long-awaited approval of spot Bitcoin exchange-traded funds (ETFs).\u0000This innovation provides investors with a new, regulated path to gain exposure\u0000to Bitcoin through a familiar investment vehicle (Kumar et al., 2024). However,\u0000unlike traditional ETFs that directly hold underlying assets, Bitcoin ETFs rely\u0000on a creation and redemption process managed by authorized participants (APs).\u0000This unique structure introduces distinct characteristics in terms of\u0000premium/discount behavior compared to traditional ETFs. This paper investigates\u0000the premium and discount patterns observed in Bitcoin ETFs during first\u0000four-month period (January 11th, 2024, to May 17th, 2024). Our analysis reveals\u0000that these patterns differ significantly from those observed in traditional\u0000index ETFs, potentially exposing investors to additional risk factors. By\u0000identifying and analyzing these risk factors associated with Bitcoin ETF\u0000premiums/discounts, this paper aims to achieve two key objectives: Enhance\u0000market understanding: Equip and market and investors with a deeper\u0000comprehension of the unique liquidity risks inherent in Bitcoin ETFs. Provide a\u0000clearer risk management frameworks: Offer a clearer perspective on the\u0000risk-return profile of digital asset ETFs, specifically focusing on Bitcoin\u0000ETFs. Through a thorough analysis of premium/discount behavior and the\u0000underlying factors contributing to it, this paper strives to contribute\u0000valuable insights for investors navigating the evolving landscape of digital\u0000asset investments","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Order flow modeling stands as the most fundamental and essential financial task, as orders embody the minimal unit within a financial market. However, current approaches often result in unsatisfactory fidelity in generating order flow, and their generation lacks controllability, thereby limiting their application scenario. In this paper, we advocate incorporating controllability into the market generation process, and propose a Diffusion Guided meta Agent(DiGA) model to address the problem. Specifically, we utilize a diffusion model to capture dynamics of market state represented by time-evolving distribution parameters about mid-price return rate and order arrival rate, and define a meta agent with financial economic priors to generate orders from the corresponding distributions. Extensive experimental results demonstrate that our method exhibits outstanding controllability and fidelity in generation. Furthermore, we validate DiGA's effectiveness as generative environment for downstream financial applications.
{"title":"Controllable Financial Market Generation with Diffusion Guided Meta Agent","authors":"Yu-Hao Huang, Chang Xu, Yang Liu, Weiqing Liu, Wu-Jun Li, Jiang Bian","doi":"arxiv-2408.12991","DOIUrl":"https://doi.org/arxiv-2408.12991","url":null,"abstract":"Order flow modeling stands as the most fundamental and essential financial\u0000task, as orders embody the minimal unit within a financial market. However,\u0000current approaches often result in unsatisfactory fidelity in generating order\u0000flow, and their generation lacks controllability, thereby limiting their\u0000application scenario. In this paper, we advocate incorporating controllability\u0000into the market generation process, and propose a Diffusion Guided meta\u0000Agent(DiGA) model to address the problem. Specifically, we utilize a diffusion\u0000model to capture dynamics of market state represented by time-evolving\u0000distribution parameters about mid-price return rate and order arrival rate, and\u0000define a meta agent with financial economic priors to generate orders from the\u0000corresponding distributions. Extensive experimental results demonstrate that\u0000our method exhibits outstanding controllability and fidelity in generation.\u0000Furthermore, we validate DiGA's effectiveness as generative environment for\u0000downstream financial applications.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We provide an economic model of Execution Tickets and use it to study the ability of the Ethereum protocol to capture MEV from block construction. We demonstrate that Execution Tickets extract all MEV when all buyers are homogeneous, risk neutral and face no capital costs. We also show that MEV capture decreases with risk aversion and capital costs. Moreover, when buyers are heterogeneous, MEV capture can be especially low and a single dominant buyer can extract much of the MEV. This adverse effect can be partially mitigated by the presence of a Proposer Builder Separation (PBS) mechanism, which gives ET buyers access to a market of specialized builders, but in practice centralization vectors still persist. With PBS, ETs are concentrated among those with the highest ex-ante MEV extraction ability and lowest cost of capital. We show how it is possible that large investors that are not builders but have substantial advantage in capital cost can come to dominate the ET market.
我们提供了执行票据的经济模型，并用它来研究以太坊协议从区块构建中获取 MEV 的可能性。我们证明，当所有买家都是同质的、风险中性且没有资本成本时，执行票据可以提取所有 MEV。我们还证明，MEV 捕获量会随着风险规避和资本成本的增加而减少。此外，当买方是异质的时候，MEV 捕获量会特别低，一个占主导地位的买方可以提取大部分的 MEV。这种不利影响可以通过建议者与建造者分离（PBS）机制得到部分缓解，该机制使 ET 购买者能够进入专业建造者市场，但在实践中，集中化矢量仍然存在。有了 PBS，ET 就会集中在那些事前提取 MEV 能力最强、资本成本最低的企业中。我们展示了并非建筑商但在资本成本方面具有巨大优势的大型投资者是如何主导 ET 市场的。
{"title":"MEV Capture and Decentralization in Execution Tickets","authors":"Jonah Burian, Davide Crapis, Fahad Saleh","doi":"arxiv-2408.11255","DOIUrl":"https://doi.org/arxiv-2408.11255","url":null,"abstract":"We provide an economic model of Execution Tickets and use it to study the\u0000ability of the Ethereum protocol to capture MEV from block construction. We\u0000demonstrate that Execution Tickets extract all MEV when all buyers are\u0000homogeneous, risk neutral and face no capital costs. We also show that MEV\u0000capture decreases with risk aversion and capital costs. Moreover, when buyers\u0000are heterogeneous, MEV capture can be especially low and a single dominant\u0000buyer can extract much of the MEV. This adverse effect can be partially\u0000mitigated by the presence of a Proposer Builder Separation (PBS) mechanism,\u0000which gives ET buyers access to a market of specialized builders, but in\u0000practice centralization vectors still persist. With PBS, ETs are concentrated\u0000among those with the highest ex-ante MEV extraction ability and lowest cost of\u0000capital. We show how it is possible that large investors that are not builders\u0000but have substantial advantage in capital cost can come to dominate the ET\u0000market.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper we introduce a multi-agent deep-learning method which trades in the Futures markets based on the US S&P 500 index. The method (referred to as Model A) is an innovation founded on existing well-established machine-learning models which sample market prices and associated derivatives in order to decide whether the investment should be long/short or closed (zero exposure), on a day-to-day decision. We compare the predictions with some conventional machine-learning methods namely, Long Short-Term Memory, Random Forest and Gradient-Boosted-Trees. Results are benchmarked against a passive model in which the Futures contracts are held (long) continuously with the same exposure (level of investment). Historical tests are based on daily daytime trading carried out over a period of 6 calendar years (2018-23). We find that Model A outperforms the passive investment in key performance metrics, placing it within the top quartile performance of US Large Cap active fund managers. Model A also outperforms the three machine-learning classification comparators over this period. We observe that Model A is extremely efficient (doing less and getting more) with an exposure to the market of only 41.95% compared to the 100% market exposure of the passive investment, and thus provides increased profitability with reduced risk.
在本文中，我们介绍了一种基于美国标准普尔 500 指数在期货市场上进行交易的多代理深度学习方法。该方法（称为模型 A）是在现有成熟的机器学习模型基础上的创新，这些模型对市场价格和相关衍生品进行采样，以决定投资是做多/做空还是平仓（零风险敞口）。我们将预测结果与一些传统的机器学习方法（即长短期记忆、随机森林和梯度增强树）进行了比较。结果以被动模型为基准，在被动模型中，期货合约以相同的风险敞口（投资水平）持续持有（做多）。历史测试基于 6 个日历年（2018-23 年）期间进行的每日日间交易。我们发现，模型 A 在关键绩效指标上的表现优于被动投资，在美国大盘股主动基金经理中名列前四分之一。在此期间，模型 A 的表现也优于三个机器学习分类比较对象。我们发现，与被动投资的 100% 市场风险敞口相比，模型 A 的市场风险敞口仅为 41.95%，具有极高的效率（少做多得），因此在降低风险的同时提高了盈利能力。
{"title":"Less is more: AI Decision-Making using Dynamic Deep Neural Networks for Short-Term Stock Index Prediction","authors":"CJ Finnegan, James F. McCann, Salissou Moutari","doi":"arxiv-2408.11740","DOIUrl":"https://doi.org/arxiv-2408.11740","url":null,"abstract":"In this paper we introduce a multi-agent deep-learning method which trades in\u0000the Futures markets based on the US S&P 500 index. The method (referred to as\u0000Model A) is an innovation founded on existing well-established machine-learning\u0000models which sample market prices and associated derivatives in order to decide\u0000whether the investment should be long/short or closed (zero exposure), on a\u0000day-to-day decision. We compare the predictions with some conventional\u0000machine-learning methods namely, Long Short-Term Memory, Random Forest and\u0000Gradient-Boosted-Trees. Results are benchmarked against a passive model in\u0000which the Futures contracts are held (long) continuously with the same exposure\u0000(level of investment). Historical tests are based on daily daytime trading\u0000carried out over a period of 6 calendar years (2018-23). We find that Model A\u0000outperforms the passive investment in key performance metrics, placing it\u0000within the top quartile performance of US Large Cap active fund managers. Model\u0000A also outperforms the three machine-learning classification comparators over\u0000this period. We observe that Model A is extremely efficient (doing less and\u0000getting more) with an exposure to the market of only 41.95% compared to the\u0000100% market exposure of the passive investment, and thus provides increased\u0000profitability with reduced risk.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sid Bhatia, Sidharth Peri, Sam Friedman, Michelle Malen
This research presents a comprehensive framework for analyzing liquidity in financial markets, particularly in the context of high-frequency trading. By leveraging advanced machine learning classification techniques, including Logistic Regression, Support Vector Machine, and Random Forest, the study aims to predict minute-level price movements using an extensive set of liquidity metrics derived from the Trade and Quote (TAQ) data. The findings reveal that employing a broad spectrum of liquidity measures yields higher predictive accuracy compared to models utilizing a reduced subset of features. Key liquidity metrics, such as Liquidity Ratio, Flow Ratio, and Turnover, consistently emerged as significant predictors across all models, with the Random Forest algorithm demonstrating superior accuracy. This study not only underscores the critical role of liquidity in market stability and transaction costs but also highlights the complexities involved in short-interval market predictions. The research suggests that a comprehensive set of liquidity measures is essential for accurate prediction, and proposes future work to validate these findings across different stock datasets to assess their generalizability.
{"title":"High-Frequency Trading Liquidity Analysis | Application of Machine Learning Classification","authors":"Sid Bhatia, Sidharth Peri, Sam Friedman, Michelle Malen","doi":"arxiv-2408.10016","DOIUrl":"https://doi.org/arxiv-2408.10016","url":null,"abstract":"This research presents a comprehensive framework for analyzing liquidity in\u0000financial markets, particularly in the context of high-frequency trading. By\u0000leveraging advanced machine learning classification techniques, including\u0000Logistic Regression, Support Vector Machine, and Random Forest, the study aims\u0000to predict minute-level price movements using an extensive set of liquidity\u0000metrics derived from the Trade and Quote (TAQ) data. The findings reveal that\u0000employing a broad spectrum of liquidity measures yields higher predictive\u0000accuracy compared to models utilizing a reduced subset of features. Key\u0000liquidity metrics, such as Liquidity Ratio, Flow Ratio, and Turnover,\u0000consistently emerged as significant predictors across all models, with the\u0000Random Forest algorithm demonstrating superior accuracy. This study not only\u0000underscores the critical role of liquidity in market stability and transaction\u0000costs but also highlights the complexities involved in short-interval market\u0000predictions. The research suggests that a comprehensive set of liquidity\u0000measures is essential for accurate prediction, and proposes future work to\u0000validate these findings across different stock datasets to assess their\u0000generalizability.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：