This paper explores the effectiveness of high-frequency options trading strategies enhanced by advanced portfolio optimization techniques, investigating their ability to consistently generate positive returns compared to traditional long or short positions on options. Utilizing SPY options data recorded in five-minute intervals over a one-month period, we calculate key metrics such as Option Greeks and implied volatility, applying the Binomial Tree model for American options pricing and the Newton-Raphson algorithm for implied volatility calculation. Investment universes are constructed based on criteria like implied volatility and Greeks, followed by the application of various portfolio optimization models, including Standard Mean-Variance and Robust Methods. Our research finds that while basic long-short strategies centered on implied volatility and Greeks generally underperform, more sophisticated strategies incorporating advanced Greeks, such as Vega and Rho, along with dynamic portfolio optimization, show potential in effectively navigating the complexities of the options market. The study highlights the importance of adaptability and responsiveness in dynamic portfolio strategies within the high-frequency trading environment, particularly under volatile market conditions. Future research could refine strategy parameters and explore less frequently traded options, offering new insights into high-frequency options trading and portfolio management.
{"title":"High-Frequency Options Trading | With Portfolio Optimization","authors":"Sid Bhatia","doi":"arxiv-2408.08866","DOIUrl":"https://doi.org/arxiv-2408.08866","url":null,"abstract":"This paper explores the effectiveness of high-frequency options trading\u0000strategies enhanced by advanced portfolio optimization techniques,\u0000investigating their ability to consistently generate positive returns compared\u0000to traditional long or short positions on options. Utilizing SPY options data\u0000recorded in five-minute intervals over a one-month period, we calculate key\u0000metrics such as Option Greeks and implied volatility, applying the Binomial\u0000Tree model for American options pricing and the Newton-Raphson algorithm for\u0000implied volatility calculation. Investment universes are constructed based on\u0000criteria like implied volatility and Greeks, followed by the application of\u0000various portfolio optimization models, including Standard Mean-Variance and\u0000Robust Methods. Our research finds that while basic long-short strategies\u0000centered on implied volatility and Greeks generally underperform, more\u0000sophisticated strategies incorporating advanced Greeks, such as Vega and Rho,\u0000along with dynamic portfolio optimization, show potential in effectively\u0000navigating the complexities of the options market. The study highlights the\u0000importance of adaptability and responsiveness in dynamic portfolio strategies\u0000within the high-frequency trading environment, particularly under volatile\u0000market conditions. Future research could refine strategy parameters and explore\u0000less frequently traded options, offering new insights into high-frequency\u0000options trading and portfolio management.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"63 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194505","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}
Market information events are generated intermittently and disseminated at high speeds in real-time. Market participants consume this high-frequency data to build limit order books, representing the current bids and offers for a given asset. The arrival processes, or the order flow of bid and offer events, are asymmetric and possibly dependent on each other. The quantum and direction of this asymmetry are often associated with the direction of the traded price movement. The Order Flow Imbalance (OFI) is an indicator commonly used to estimate this asymmetry. This paper uses Hawkes processes to estimate the OFI while accounting for the lagged dependence in the order flow between bids and offers. Secondly, we develop a method to forecast the near-term distribution of the OFI, which can then be used to compare models for forecasting OFI. Thirdly, we propose a method to compare the forecasts of OFI for an arbitrarily large number of models. We apply the approach developed to tick data from the National Stock Exchange and observe that the Hawkes process modeled with a Sum of Exponential's kernel gives the best forecast among all competing models.
{"title":"Forecasting High Frequency Order Flow Imbalance","authors":"Aditya Nittur Anantha, Shashi Jain","doi":"arxiv-2408.03594","DOIUrl":"https://doi.org/arxiv-2408.03594","url":null,"abstract":"Market information events are generated intermittently and disseminated at\u0000high speeds in real-time. Market participants consume this high-frequency data\u0000to build limit order books, representing the current bids and offers for a\u0000given asset. The arrival processes, or the order flow of bid and offer events,\u0000are asymmetric and possibly dependent on each other. The quantum and direction\u0000of this asymmetry are often associated with the direction of the traded price\u0000movement. The Order Flow Imbalance (OFI) is an indicator commonly used to\u0000estimate this asymmetry. This paper uses Hawkes processes to estimate the OFI\u0000while accounting for the lagged dependence in the order flow between bids and\u0000offers. Secondly, we develop a method to forecast the near-term distribution of\u0000the OFI, which can then be used to compare models for forecasting OFI. Thirdly,\u0000we propose a method to compare the forecasts of OFI for an arbitrarily large\u0000number of models. We apply the approach developed to tick data from the\u0000National Stock Exchange and observe that the Hawkes process modeled with a Sum\u0000of Exponential's kernel gives the best forecast among all competing models.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"30 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141948368","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 use random walks to simulate the fluid limit of two coupled diffusive limit order books to model correlation emergence. The model implements the arrival, cancellation and diffusion of orders coupled by a pairs trader profiting from the mean-reversion between the two order books in the fluid limit for a Lit order book with vanishing boundary conditions and order volume conservation. We are able to demonstrate the recovery of an Epps effect from this. We discuss how various stylised facts depend on the model parameters and the numerical scheme and discuss the various strengths and weaknesses of the approach. We demonstrate how the Epps effect depends on different choices of time and price discretisation. This shows how an Epps effect can emerge without recourse to market microstructure noise relative to a latent model but can rather be viewed as an emergent property arising from trader interactions in a world of asynchronous events.
{"title":"Correlation emergence in two coupled simulated limit order books","authors":"Dominic Bauer, Derick Diana, Tim Gebbie","doi":"arxiv-2408.03181","DOIUrl":"https://doi.org/arxiv-2408.03181","url":null,"abstract":"We use random walks to simulate the fluid limit of two coupled diffusive\u0000limit order books to model correlation emergence. The model implements the\u0000arrival, cancellation and diffusion of orders coupled by a pairs trader\u0000profiting from the mean-reversion between the two order books in the fluid\u0000limit for a Lit order book with vanishing boundary conditions and order volume\u0000conservation. We are able to demonstrate the recovery of an Epps effect from\u0000this. We discuss how various stylised facts depend on the model parameters and\u0000the numerical scheme and discuss the various strengths and weaknesses of the\u0000approach. We demonstrate how the Epps effect depends on different choices of\u0000time and price discretisation. This shows how an Epps effect can emerge without\u0000recourse to market microstructure noise relative to a latent model but can\u0000rather be viewed as an emergent property arising from trader interactions in a\u0000world of asynchronous events.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141948370","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}
Reinforcement learning works best when the impact of the agent's actions on its environment can be perfectly simulated or fully appraised from available data. Some systems are however both hard to simulate and very sensitive to small perturbations. An additional difficulty arises when an RL agent must learn to be part of a multi-agent system using only anonymous data, which makes it impossible to infer the state of each agent, thus to use data directly. Typical examples are competitive systems without agent-resolved data such as financial markets. We introduce consistent data time travel for offline RL as a remedy for these problems: instead of using historical data in a sequential way, we argue that one needs to perform time travel in historical data, i.e., to adjust the time index so that both the past state and the influence of the RL agent's action on the state coincide with real data. This both alleviates the need to resort to imperfect models and consistently accounts for both the immediate and long-term reactions of the system when using anonymous historical data. We apply this idea to market making in limit order books, a notoriously difficult task for RL; it turns out that the gain of the agent is significantly higher with data time travel than with naive sequential data, which suggests that the difficulty of this task for RL may have been overestimated.
{"title":"Data time travel and consistent market making: taming reinforcement learning in multi-agent systems with anonymous data","authors":"Vincent Ragel, Damien Challet","doi":"arxiv-2408.02322","DOIUrl":"https://doi.org/arxiv-2408.02322","url":null,"abstract":"Reinforcement learning works best when the impact of the agent's actions on\u0000its environment can be perfectly simulated or fully appraised from available\u0000data. Some systems are however both hard to simulate and very sensitive to\u0000small perturbations. An additional difficulty arises when an RL agent must\u0000learn to be part of a multi-agent system using only anonymous data, which makes\u0000it impossible to infer the state of each agent, thus to use data directly.\u0000Typical examples are competitive systems without agent-resolved data such as\u0000financial markets. We introduce consistent data time travel for offline RL as a\u0000remedy for these problems: instead of using historical data in a sequential\u0000way, we argue that one needs to perform time travel in historical data, i.e.,\u0000to adjust the time index so that both the past state and the influence of the\u0000RL agent's action on the state coincide with real data. This both alleviates\u0000the need to resort to imperfect models and consistently accounts for both the\u0000immediate and long-term reactions of the system when using anonymous historical\u0000data. We apply this idea to market making in limit order books, a notoriously\u0000difficult task for RL; it turns out that the gain of the agent is significantly\u0000higher with data time travel than with naive sequential data, which suggests\u0000that the difficulty of this task for RL may have been overestimated.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"45 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141948369","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}
Robert McLaughlin, Nir Chemaya, Dingyue Liu, Dahlia Malkhi
Trading on decentralized exchanges via an Automated Market Maker (AMM) mechanism has been massively adopted, with a daily trading volume reaching $1B. This trading method has also received close attention from researchers, central banks, and financial firms, who have the potential to adopt it to traditional financial markets such as foreign exchanges and stock markets. A critical challenge of AMM-powered trading is that transaction order has high financial value, so a policy or method to order transactions in a "good" (optimal) manner is vital. We offer economic measures of both price stability (low volatility) and inequality that inform how a "social planner" should pick an optimal ordering. We show that there is a trade-off between achieving price stability and reducing inequality, and that policymakers must choose which to prioritize. In addition, picking the optimal order can often be costly, especially when performing an exhaustive search over trade orderings (permutations). As an alternative we provide a simple algorithm, Clever Look-ahead Volatility Reduction (CLVR). This algorithm constructs an ordering which approximately minimizes price volatility with a small computation cost. We also provide insight into the strategy changes that may occur if traders are subject to this sequencing algorithm.
{"title":"CLVR Ordering of Transactions on AMMs","authors":"Robert McLaughlin, Nir Chemaya, Dingyue Liu, Dahlia Malkhi","doi":"arxiv-2408.02634","DOIUrl":"https://doi.org/arxiv-2408.02634","url":null,"abstract":"Trading on decentralized exchanges via an Automated Market Maker (AMM)\u0000mechanism has been massively adopted, with a daily trading volume reaching $1B.\u0000This trading method has also received close attention from researchers, central\u0000banks, and financial firms, who have the potential to adopt it to traditional\u0000financial markets such as foreign exchanges and stock markets. A critical\u0000challenge of AMM-powered trading is that transaction order has high financial\u0000value, so a policy or method to order transactions in a \"good\" (optimal) manner\u0000is vital. We offer economic measures of both price stability (low volatility)\u0000and inequality that inform how a \"social planner\" should pick an optimal\u0000ordering. We show that there is a trade-off between achieving price stability\u0000and reducing inequality, and that policymakers must choose which to prioritize.\u0000In addition, picking the optimal order can often be costly, especially when\u0000performing an exhaustive search over trade orderings (permutations). As an\u0000alternative we provide a simple algorithm, Clever Look-ahead Volatility\u0000Reduction (CLVR). This algorithm constructs an ordering which approximately\u0000minimizes price volatility with a small computation cost. We also provide\u0000insight into the strategy changes that may occur if traders are subject to this\u0000sequencing algorithm.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"64 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141948411","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 introduce a novel approach to options trading strategies using a highly scalable and data-driven machine learning algorithm. In contrast to traditional approaches that often require specifications of underlying market dynamics or assumptions on an option pricing model, our models depart fundamentally from the need for these prerequisites, directly learning non-trivial mappings from market data to optimal trading signals. Backtesting on more than a decade of option contracts for equities listed on the S&P 100, we demonstrate that deep learning models trained according to our end-to-end approach exhibit significant improvements in risk-adjusted performance over existing rules-based trading strategies. We find that incorporating turnover regularization into the models leads to further performance enhancements at prohibitively high levels of transaction costs.
{"title":"Deep Learning for Options Trading: An End-To-End Approach","authors":"Wee Ling Tan, Stephen Roberts, Stefan Zohren","doi":"arxiv-2407.21791","DOIUrl":"https://doi.org/arxiv-2407.21791","url":null,"abstract":"We introduce a novel approach to options trading strategies using a highly\u0000scalable and data-driven machine learning algorithm. In contrast to traditional\u0000approaches that often require specifications of underlying market dynamics or\u0000assumptions on an option pricing model, our models depart fundamentally from\u0000the need for these prerequisites, directly learning non-trivial mappings from\u0000market data to optimal trading signals. Backtesting on more than a decade of\u0000option contracts for equities listed on the S&P 100, we demonstrate that deep\u0000learning models trained according to our end-to-end approach exhibit\u0000significant improvements in risk-adjusted performance over existing rules-based\u0000trading strategies. We find that incorporating turnover regularization into the\u0000models leads to further performance enhancements at prohibitively high levels\u0000of transaction costs.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"21 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141873179","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}
Trading is a highly competitive task that requires a combination of strategy, knowledge, and psychological fortitude. With the recent success of large language models(LLMs), it is appealing to apply the emerging intelligence of LLM agents in this competitive arena and understanding if they can outperform professional traders. In this survey, we provide a comprehensive review of the current research on using LLMs as agents in financial trading. We summarize the common architecture used in the agent, the data inputs, and the performance of LLM trading agents in backtesting as well as the challenges presented in these research. This survey aims to provide insights into the current state of LLM-based financial trading agents and outline future research directions in this field.
{"title":"Large Language Model Agent in Financial Trading: A Survey","authors":"Han Ding, Yinheng Li, Junhao Wang, Hang Chen","doi":"arxiv-2408.06361","DOIUrl":"https://doi.org/arxiv-2408.06361","url":null,"abstract":"Trading is a highly competitive task that requires a combination of strategy,\u0000knowledge, and psychological fortitude. With the recent success of large\u0000language models(LLMs), it is appealing to apply the emerging intelligence of\u0000LLM agents in this competitive arena and understanding if they can outperform\u0000professional traders. In this survey, we provide a comprehensive review of the\u0000current research on using LLMs as agents in financial trading. We summarize the\u0000common architecture used in the agent, the data inputs, and the performance of\u0000LLM trading agents in backtesting as well as the challenges presented in these\u0000research. This survey aims to provide insights into the current state of\u0000LLM-based financial trading agents and outline future research directions in\u0000this field.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"15 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194518","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}
Market making refers to a form of trading in financial markets characterized by passive orders which add liquidity to limit order books. Market makers are important for the proper functioning of financial markets worldwide. Given the importance, financial mathematics has endeavored to derive optimal strategies for placing limit orders in this context. This paper identifies a key discrepancy between popular model assumptions and the realities of real markets, specifically regarding the dynamics around limit order fills. Traditionally, market making models rely on an assumption of low-cost random fills, when in reality we observe a high-cost non-random fill behavior. Namely, limit order fills are caused by and coincide with adverse price movements, which create a drag on the market maker's profit and loss. We refer to this phenomenon as "the negative drift" associated with limit order fills. We describe a discrete market model and prove theoretically that the negative drift exists. We also provide a detailed empirical simulation using one of the most traded financial instruments in the world, the 10 Year US Treasury Bond futures, which also confirms its existence. To our knowledge, this is the first paper to describe and prove this phenomenon in such detail.
{"title":"The Negative Drift of a Limit Order Fill","authors":"Timothy DeLise","doi":"arxiv-2407.16527","DOIUrl":"https://doi.org/arxiv-2407.16527","url":null,"abstract":"Market making refers to a form of trading in financial markets characterized\u0000by passive orders which add liquidity to limit order books. Market makers are\u0000important for the proper functioning of financial markets worldwide. Given the\u0000importance, financial mathematics has endeavored to derive optimal strategies\u0000for placing limit orders in this context. This paper identifies a key\u0000discrepancy between popular model assumptions and the realities of real\u0000markets, specifically regarding the dynamics around limit order fills.\u0000Traditionally, market making models rely on an assumption of low-cost random\u0000fills, when in reality we observe a high-cost non-random fill behavior. Namely,\u0000limit order fills are caused by and coincide with adverse price movements,\u0000which create a drag on the market maker's profit and loss. We refer to this\u0000phenomenon as \"the negative drift\" associated with limit order fills. We\u0000describe a discrete market model and prove theoretically that the negative\u0000drift exists. We also provide a detailed empirical simulation using one of the\u0000most traded financial instruments in the world, the 10 Year US Treasury Bond\u0000futures, which also confirms its existence. To our knowledge, this is the first\u0000paper to describe and prove this phenomenon in such detail.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"14 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141773571","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}
Automated market makers (AMMs) are a new type of trading venues which are revolutionising the way market participants interact. At present, the majority of AMMs are constant function market makers (CFMMs) where a deterministic trading function determines how markets are cleared. Within CFMMs, we focus on constant product market makers (CPMMs) which implements the concentrated liquidity (CL) feature. In this thesis we formalise and study the trading mechanism of CPMMs with CL, and we develop liquidity provision and liquidity taking strategies. Our models are motivated and tested with market data. We derive optimal strategies for liquidity takers (LTs) who trade orders of large size and execute statistical arbitrages. First, we consider an LT who trades in a CPMM with CL and uses the dynamics of prices in competing venues as market signals. We use Uniswap v3 data to study price, liquidity, and trading cost dynamics, and to motivate the model. Next, we consider an LT who trades a basket of crypto-currencies whose constituents co-move. We use market data to study lead-lag effects, spillover effects, and causality between trading venues. We derive optimal strategies for strategic liquidity providers (LPs) who provide liquidity in CPMM with CL. First, we use stochastic control tools to derive a self-financing and closed-form optimal liquidity provision strategy where the width of the LP's liquidity range is determined by the profitability of the pool, the dynamics of the LP's position, and concentration risk. Next, we use a model-free approach to solve the problem of an LP who provides liquidity in multiple CPMMs with CL. We do not specify a model for the stochastic processes observed by LPs, and use a long short-term memory (LSTM) neural network to approximate the optimal liquidity provision strategy.
{"title":"Automated Market Making and Decentralized Finance","authors":"Marcello Monga","doi":"arxiv-2407.16885","DOIUrl":"https://doi.org/arxiv-2407.16885","url":null,"abstract":"Automated market makers (AMMs) are a new type of trading venues which are\u0000revolutionising the way market participants interact. At present, the majority\u0000of AMMs are constant function market makers (CFMMs) where a deterministic\u0000trading function determines how markets are cleared. Within CFMMs, we focus on\u0000constant product market makers (CPMMs) which implements the concentrated\u0000liquidity (CL) feature. In this thesis we formalise and study the trading\u0000mechanism of CPMMs with CL, and we develop liquidity provision and liquidity\u0000taking strategies. Our models are motivated and tested with market data. We derive optimal strategies for liquidity takers (LTs) who trade orders of\u0000large size and execute statistical arbitrages. First, we consider an LT who\u0000trades in a CPMM with CL and uses the dynamics of prices in competing venues as\u0000market signals. We use Uniswap v3 data to study price, liquidity, and trading\u0000cost dynamics, and to motivate the model. Next, we consider an LT who trades a\u0000basket of crypto-currencies whose constituents co-move. We use market data to\u0000study lead-lag effects, spillover effects, and causality between trading\u0000venues. We derive optimal strategies for strategic liquidity providers (LPs) who\u0000provide liquidity in CPMM with CL. First, we use stochastic control tools to\u0000derive a self-financing and closed-form optimal liquidity provision strategy\u0000where the width of the LP's liquidity range is determined by the profitability\u0000of the pool, the dynamics of the LP's position, and concentration risk. Next,\u0000we use a model-free approach to solve the problem of an LP who provides\u0000liquidity in multiple CPMMs with CL. We do not specify a model for the\u0000stochastic processes observed by LPs, and use a long short-term memory (LSTM)\u0000neural network to approximate the optimal liquidity provision strategy.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"13 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141773568","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}
Cryptocurrency is a cryptography-based digital asset with extremely volatile prices. Around $70 billion worth of crypto-currency is traded daily on exchanges. Trading crypto-currency is difficult due to the inherent volatility of the crypto-market. In this work, we want to test the hypothesis: "Can techniques from artificial intelligence help with algorithmically trading cryptocurrencies?". In order to address this question, we combine Reinforcement Learning (RL) with pair trading. Pair trading is a statistical arbitrage trading technique which exploits the price difference between statistically correlated assets. We train reinforcement learners to determine when and how to trade pairs of cryptocurrencies. We develop new reward shaping and observation/action spaces for reinforcement learning. We performed experiments with the developed reinforcement learner on pairs of BTC-GBP and BTC-EUR data separated by 1-minute intervals (n = 263,520). The traditional non-RL pair trading technique achieved an annualised profit of 8.33%, while the proposed RL-based pair trading technique achieved annualised profits from 9.94% - 31.53%, depending upon the RL learner. Our results show that RL can significantly outperform manual and traditional pair trading techniques when applied to volatile markets such as cryptocurrencies.
{"title":"Reinforcement Learning Pair Trading: A Dynamic Scaling approach","authors":"Hongshen Yang, Avinash Malik","doi":"arxiv-2407.16103","DOIUrl":"https://doi.org/arxiv-2407.16103","url":null,"abstract":"Cryptocurrency is a cryptography-based digital asset with extremely volatile\u0000prices. Around $70 billion worth of crypto-currency is traded daily on\u0000exchanges. Trading crypto-currency is difficult due to the inherent volatility\u0000of the crypto-market. In this work, we want to test the hypothesis: \"Can\u0000techniques from artificial intelligence help with algorithmically trading\u0000cryptocurrencies?\". In order to address this question, we combine Reinforcement\u0000Learning (RL) with pair trading. Pair trading is a statistical arbitrage\u0000trading technique which exploits the price difference between statistically\u0000correlated assets. We train reinforcement learners to determine when and how to\u0000trade pairs of cryptocurrencies. We develop new reward shaping and\u0000observation/action spaces for reinforcement learning. We performed experiments\u0000with the developed reinforcement learner on pairs of BTC-GBP and BTC-EUR data\u0000separated by 1-minute intervals (n = 263,520). The traditional non-RL pair\u0000trading technique achieved an annualised profit of 8.33%, while the proposed\u0000RL-based pair trading technique achieved annualised profits from 9.94% -\u000031.53%, depending upon the RL learner. Our results show that RL can\u0000significantly outperform manual and traditional pair trading techniques when\u0000applied to volatile markets such as cryptocurrencies.","PeriodicalId":501478,"journal":{"name":"arXiv - QuantFin - Trading and Market Microstructure","volume":"38 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141773629","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：