This study proposes an innovative evaluation method based on large language models (LLMs) specifically designed to measure the digital transformation (DT) process of enterprises. By analyzing the annual reports of 4407 companies listed on the New York Stock Exchange and Nasdaq from 2005 to 2022, a comprehensive set of DT indicators was constructed. The findings revealed that DT significantly improves a company's financial performance, however, different digital technologies exhibit varying effects on financial performance. Specifically, blockchain technology has a relatively limited positive impact on financial performance. In addition, this study further discovered that DT can promote the growth of financial performance by enhancing operational efficiency and reducing costs. This study provides a novel DT evaluation tool for the academic community, while also expanding the application scope of generative artificial intelligence technology in economic research.
{"title":"New intelligent empowerment for digital transformation","authors":"Peng Yifeng, Gao Chen","doi":"arxiv-2406.18440","DOIUrl":"https://doi.org/arxiv-2406.18440","url":null,"abstract":"This study proposes an innovative evaluation method based on large language\u0000models (LLMs) specifically designed to measure the digital transformation (DT)\u0000process of enterprises. By analyzing the annual reports of 4407 companies\u0000listed on the New York Stock Exchange and Nasdaq from 2005 to 2022, a\u0000comprehensive set of DT indicators was constructed. The findings revealed that\u0000DT significantly improves a company's financial performance, however, different\u0000digital technologies exhibit varying effects on financial performance.\u0000Specifically, blockchain technology has a relatively limited positive impact on\u0000financial performance. In addition, this study further discovered that DT can\u0000promote the growth of financial performance by enhancing operational efficiency\u0000and reducing costs. This study provides a novel DT evaluation tool for the\u0000academic community, while also expanding the application scope of generative\u0000artificial intelligence technology in economic research.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"29 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141501635","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}
Alphas are pivotal in providing signals for quantitative trading. The industry highly values the discovery of formulaic alphas for their interpretability and ease of analysis, compared with the expressive yet overfitting-prone black-box alphas. In this work, we focus on discovering formulaic alphas. Prior studies on automatically generating a collection of formulaic alphas were mostly based on genetic programming (GP), which is known to suffer from the problems of being sensitive to the initial population, converting to local optima, and slow computation speed. Recent efforts employing deep reinforcement learning (DRL) for alpha discovery have not fully addressed key practical considerations such as alpha correlations and validity, which are crucial for their effectiveness. In this work, we propose a novel framework for alpha discovery using DRL by formulating the alpha discovery process as program construction. Our agent, $text{Alpha}^2$, assembles an alpha program optimized for an evaluation metric. A search algorithm guided by DRL navigates through the search space based on value estimates for potential alpha outcomes. The evaluation metric encourages both the performance and the diversity of alphas for a better final trading strategy. Our formulation of searching alphas also brings the advantage of pre-calculation dimensional analysis, ensuring the logical soundness of alphas, and pruning the vast search space to a large extent. Empirical experiments on real-world stock markets demonstrates $text{Alpha}^2$'s capability to identify a diverse set of logical and effective alphas, which significantly improves the performance of the final trading strategy. The code of our method is available at https://github.com/x35f/alpha2.
{"title":"$text{Alpha}^2$: Discovering Logical Formulaic Alphas using Deep Reinforcement Learning","authors":"Feng Xu, Yan Yin, Xinyu Zhang, Tianyuan Liu, Shengyi Jiang, Zongzhang Zhang","doi":"arxiv-2406.16505","DOIUrl":"https://doi.org/arxiv-2406.16505","url":null,"abstract":"Alphas are pivotal in providing signals for quantitative trading. The\u0000industry highly values the discovery of formulaic alphas for their\u0000interpretability and ease of analysis, compared with the expressive yet\u0000overfitting-prone black-box alphas. In this work, we focus on discovering\u0000formulaic alphas. Prior studies on automatically generating a collection of\u0000formulaic alphas were mostly based on genetic programming (GP), which is known\u0000to suffer from the problems of being sensitive to the initial population,\u0000converting to local optima, and slow computation speed. Recent efforts\u0000employing deep reinforcement learning (DRL) for alpha discovery have not fully\u0000addressed key practical considerations such as alpha correlations and validity,\u0000which are crucial for their effectiveness. In this work, we propose a novel\u0000framework for alpha discovery using DRL by formulating the alpha discovery\u0000process as program construction. Our agent, $text{Alpha}^2$, assembles an\u0000alpha program optimized for an evaluation metric. A search algorithm guided by\u0000DRL navigates through the search space based on value estimates for potential\u0000alpha outcomes. The evaluation metric encourages both the performance and the\u0000diversity of alphas for a better final trading strategy. Our formulation of\u0000searching alphas also brings the advantage of pre-calculation dimensional\u0000analysis, ensuring the logical soundness of alphas, and pruning the vast search\u0000space to a large extent. Empirical experiments on real-world stock markets\u0000demonstrates $text{Alpha}^2$'s capability to identify a diverse set of logical\u0000and effective alphas, which significantly improves the performance of the final\u0000trading strategy. The code of our method is available at\u0000https://github.com/x35f/alpha2.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"33 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141520788","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}
Cyclic arbitrage chances exist abundantly among decentralized exchanges (DEXs), like Uniswap V2. For an arbitrage cycle (loop), researchers or practitioners usually choose a specific token, such as Ether as input, and optimize their input amount to get the net maximal amount of the specific token as arbitrage profit. By considering the tokens' prices from CEXs in this paper, the new arbitrage profit, called monetized arbitrage profit, will be quantified as the product of the net number of a specific token we got from the arbitrage loop and its corresponding price in CEXs. Based on this concept, we put forward three different strategies to maximize the monetized arbitrage profit for each arbitrage loop. The first strategy is called the MaxPrice strategy. Under this strategy, arbitrageurs start arbitrage only from the token with the highest CEX price. The second strategy is called the MaxMax strategy. Under this strategy, we calculate the monetized arbitrage profit for each token as input in turn in the arbitrage loop. Then, we pick up the most maximal monetized arbitrage profit among them as the monetized arbitrage profit of the MaxMax strategy. The third one is called the Convex Optimization strategy. By mapping the MaxMax strategy to a convex optimization problem, we proved that the Convex Optimization strategy could get more profit in theory than the MaxMax strategy, which is proved again in a given example. We also proved that if no arbitrage profit exists according to the MaxMax strategy, then the Convex Optimization strategy can not detect any arbitrage profit, either. However, the empirical data analysis denotes that the profitability of the Convex Optimization strategy is almost equal to that of the MaxMax strategy, and the MaxPrice strategy is not reliable in getting the maximal monetized arbitrage profit compared to the MaxMax strategy.
{"title":"Profit Maximization In Arbitrage Loops","authors":"Yu Zhang, Zichen Li, Tao Yan, Qianyu Liu, Nicolo Vallarano, Claudio Tessone","doi":"arxiv-2406.16600","DOIUrl":"https://doi.org/arxiv-2406.16600","url":null,"abstract":"Cyclic arbitrage chances exist abundantly among decentralized exchanges\u0000(DEXs), like Uniswap V2. For an arbitrage cycle (loop), researchers or\u0000practitioners usually choose a specific token, such as Ether as input, and\u0000optimize their input amount to get the net maximal amount of the specific token\u0000as arbitrage profit. By considering the tokens' prices from CEXs in this paper,\u0000the new arbitrage profit, called monetized arbitrage profit, will be quantified\u0000as the product of the net number of a specific token we got from the arbitrage\u0000loop and its corresponding price in CEXs. Based on this concept, we put forward\u0000three different strategies to maximize the monetized arbitrage profit for each\u0000arbitrage loop. The first strategy is called the MaxPrice strategy. Under this\u0000strategy, arbitrageurs start arbitrage only from the token with the highest CEX\u0000price. The second strategy is called the MaxMax strategy. Under this strategy,\u0000we calculate the monetized arbitrage profit for each token as input in turn in\u0000the arbitrage loop. Then, we pick up the most maximal monetized arbitrage\u0000profit among them as the monetized arbitrage profit of the MaxMax strategy. The\u0000third one is called the Convex Optimization strategy. By mapping the MaxMax\u0000strategy to a convex optimization problem, we proved that the Convex\u0000Optimization strategy could get more profit in theory than the MaxMax strategy,\u0000which is proved again in a given example. We also proved that if no arbitrage\u0000profit exists according to the MaxMax strategy, then the Convex Optimization\u0000strategy can not detect any arbitrage profit, either. However, the empirical\u0000data analysis denotes that the profitability of the Convex Optimization\u0000strategy is almost equal to that of the MaxMax strategy, and the MaxPrice\u0000strategy is not reliable in getting the maximal monetized arbitrage profit\u0000compared to the MaxMax strategy.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141501639","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}
Yu Zhang, Tao Yan, Jianhong Lin, Benjamin Kraner, Claudio Tessone
In decentralized exchanges (DEXs), the arbitrage paths exist abundantly in the form of both arbitrage loops (e.g. the arbitrage path starts from token A and back to token A again in the end, A, B,..., A) and non-loops (e.g. the arbitrage path starts from token A and stops at a different token N, A, B,..., N). The Moore-Bellman-Ford algorithm, often coupled with the ``walk to the root" technique, is commonly employed for detecting arbitrage loops in the token graph of decentralized exchanges (DEXs) such as Uniswap. However, a limitation of this algorithm is its ability to recognize only a limited number of arbitrage loops in each run. Additionally, it cannot specify the starting token of the detected arbitrage loops, further constraining its effectiveness in certain scenarios. Another limitation of this algorithm is its incapacity to detect non-loop arbitrage paths between any specified pairs of tokens. In this paper, we develop a new method to solve these problems by combining the line graph and a modified Moore-Bellman-Ford algorithm (MMBF). This method can help to find more arbitrage loops by detecting at least one arbitrage loop starting from any specified tokens in the DEXs and can detect the non-loop arbitrage paths between any pair of tokens. Then, we applied our algorithm to Uniswap V2 and found more arbitrage loops and non-loops indeed compared with applying the Moore-Bellman-Ford (MBF) combined algorithm. The found arbitrage profit by our method in some arbitrage paths can be even as high as one million dollars, far larger than that found by the MBF combined algorithm. Finally, we statistically compare the distribution of arbitrage path lengths and the arbitrage profit detected by both our method and the MBF combined algorithm, and depict how potential arbitrage opportunities change with time by our method.
在去中心化交易所（DEX）中，套利路径以套利循环（例如，套利路径从代币 A 开始，最后再次回到代币 A，A，B，...，A）和非循环（例如，套利路径从代币 A 开始，在不同的代币 N 停止，A，B，...，N）的形式大量存在。在检测 Uniswap 等去中心化交易所（DEX）的代币图中的套利循环时，通常会使用 Moore-Bellman-Ford 算法，该算法通常与 "walk to theroot "技术相结合。然而，这种算法的局限性在于每次运行只能识别有限数量的套利循环。此外，它不能指定检测到的套利循环的起始令牌，这进一步限制了它在某些情况下的有效性。该算法的另一个局限是无法检测到任何指定标记对之间的非循环套利路径。在本文中，我们结合线图和改进的摩尔-贝尔曼-福德算法（MMBF），开发了一种新方法来解决这些问题。这种方法可以通过检测从 DEXs 中任意指定令牌开始的至少一个套利环路来帮助找到更多套利环路，并且可以检测任意一对令牌之间的非环路套利路径。然后，我们将我们的算法应用于 Uniswap V2，与应用摩尔-贝尔曼-福德（MBF）组合算法相比，确实发现了更多的套利循环和非循环。在某些套利路径中，我们的方法所发现的套利利润甚至高达一百万美元，远远超过 MBF 组合算法所发现的利润。最后，我们统计比较了我们的方法和 MBF 组合算法所发现的套利路径长度分布和套利利润，并描述了我们的方法所发现的潜在套利机会是如何随时间变化的。
{"title":"An Improved Algorithm to Identify More Arbitrage Opportunities on Decentralized Exchanges","authors":"Yu Zhang, Tao Yan, Jianhong Lin, Benjamin Kraner, Claudio Tessone","doi":"arxiv-2406.16573","DOIUrl":"https://doi.org/arxiv-2406.16573","url":null,"abstract":"In decentralized exchanges (DEXs), the arbitrage paths exist abundantly in\u0000the form of both arbitrage loops (e.g. the arbitrage path starts from token A\u0000and back to token A again in the end, A, B,..., A) and non-loops (e.g. the\u0000arbitrage path starts from token A and stops at a different token N, A, B,...,\u0000N). The Moore-Bellman-Ford algorithm, often coupled with the ``walk to the\u0000root\" technique, is commonly employed for detecting arbitrage loops in the\u0000token graph of decentralized exchanges (DEXs) such as Uniswap. However, a\u0000limitation of this algorithm is its ability to recognize only a limited number\u0000of arbitrage loops in each run. Additionally, it cannot specify the starting\u0000token of the detected arbitrage loops, further constraining its effectiveness\u0000in certain scenarios. Another limitation of this algorithm is its incapacity to\u0000detect non-loop arbitrage paths between any specified pairs of tokens. In this\u0000paper, we develop a new method to solve these problems by combining the line\u0000graph and a modified Moore-Bellman-Ford algorithm (MMBF). This method can help\u0000to find more arbitrage loops by detecting at least one arbitrage loop starting\u0000from any specified tokens in the DEXs and can detect the non-loop arbitrage\u0000paths between any pair of tokens. Then, we applied our algorithm to Uniswap V2\u0000and found more arbitrage loops and non-loops indeed compared with applying the\u0000Moore-Bellman-Ford (MBF) combined algorithm. The found arbitrage profit by our\u0000method in some arbitrage paths can be even as high as one million dollars, far\u0000larger than that found by the MBF combined algorithm. Finally, we statistically\u0000compare the distribution of arbitrage path lengths and the arbitrage profit\u0000detected by both our method and the MBF combined algorithm, and depict how\u0000potential arbitrage opportunities change with time by our method.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"36 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141501640","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}
Machine learning techniques applied to the problem of financial market forecasting struggle with dynamic regime switching, or underlying correlation and covariance shifts in true (hidden) market variables. Drawing inspiration from the success of reinforcement learning in robotics, particularly in agile locomotion adaptation of quadruped robots to unseen terrains, we introduce an innovative approach that leverages world knowledge of pretrained LLMs (aka. 'privileged information' in robotics) and dynamically adapts them using intrinsic, natural market rewards using LLM alignment technique we dub as "Reinforcement Learning from Market Feedback" (**RLMF**). Strong empirical results demonstrate the efficacy of our method in adapting to regime shifts in financial markets, a challenge that has long plagued predictive models in this domain. The proposed algorithmic framework outperforms best-performing SOTA LLM models on the existing (FLARE) benchmark stock-movement (SM) tasks by more than 15% improved accuracy. On the recently proposed NIFTY SM task, our adaptive policy outperforms the SOTA best performing trillion parameter models like GPT-4. The paper details the dual-phase, teacher-student architecture and implementation of our model, the empirical results obtained, and an analysis of the role of language embeddings in terms of Information Gain.
应用于金融市场预测问题的机器学习技术在动态体制转换或真实（隐藏）市场变量的潜在相关性和协方差变化方面困难重重。我们从机器人学中强化学习的成功，特别是四足机器人对未知地形的敏捷运动适应中汲取灵感，引入了一种创新方法，即利用预训练 LLM 的世界知识（又称机器人学中的 "特权信息"），并使用我们称之为 "市场反馈强化学习"（**RLMF**）的 LLM 对齐技术，利用内在的自然市场奖励对它们进行动态调整。强大的实证结果证明了我们的方法在适应金融市场制度转变方面的功效，而这正是长期困扰该领域预测模型的难题。在现有的（FLARE）基准股票移动（SM）任务上，所提出的算法框架优于表现最好的 SOTA LLM 模型，准确率提高了 15% 以上。在最近提出的 NIFTY SM 任务中，我们的自适应策略优于 SOTA 性能最好的万亿参数模型，如 GPT-4。论文详细介绍了我们的模型的师生双阶段架构和实施、获得的实证结果以及对语言嵌入在信息增益方面的作用的分析。
{"title":"What Teaches Robots to Walk, Teaches Them to Trade too -- Regime Adaptive Execution using Informed Data and LLMs","authors":"Raeid Saqur","doi":"arxiv-2406.15508","DOIUrl":"https://doi.org/arxiv-2406.15508","url":null,"abstract":"Machine learning techniques applied to the problem of financial market\u0000forecasting struggle with dynamic regime switching, or underlying correlation\u0000and covariance shifts in true (hidden) market variables. Drawing inspiration\u0000from the success of reinforcement learning in robotics, particularly in agile\u0000locomotion adaptation of quadruped robots to unseen terrains, we introduce an\u0000innovative approach that leverages world knowledge of pretrained LLMs (aka.\u0000'privileged information' in robotics) and dynamically adapts them using\u0000intrinsic, natural market rewards using LLM alignment technique we dub as\u0000\"Reinforcement Learning from Market Feedback\" (**RLMF**). Strong empirical\u0000results demonstrate the efficacy of our method in adapting to regime shifts in\u0000financial markets, a challenge that has long plagued predictive models in this\u0000domain. The proposed algorithmic framework outperforms best-performing SOTA LLM\u0000models on the existing (FLARE) benchmark stock-movement (SM) tasks by more than\u000015% improved accuracy. On the recently proposed NIFTY SM task, our adaptive\u0000policy outperforms the SOTA best performing trillion parameter models like\u0000GPT-4. The paper details the dual-phase, teacher-student architecture and\u0000implementation of our model, the empirical results obtained, and an analysis of\u0000the role of language embeddings in terms of Information Gain.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"2012 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141520789","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}
This paper extends the possibility to examine the underlying curvature of data through the lens of topology by using the Betti curves, tools of Persistent Homology, as key topological descriptors, building on the clique topology approach. It was previously shown that Betti curves distinguish random from Euclidean geometric matrices - i.e. distance matrices of points randomly distributed in a cube with Euclidean distance. In line with previous experiments, we consider their low-dimensional approximations named integral Betti values, or signatures that effectively distinguish not only Euclidean, but also spherical and hyperbolic geometric matrices, both from purely random matrices as well as among themselves. To prove this, we analyse the behaviour of Betti curves for various geometric matrices -- i.e. distance matrices of points randomly distributed on manifolds of constant sectional curvature, considering the classical models of curvature 0, 1, -1, given by the Euclidean space, the sphere, and the hyperbolic space. We further investigate the dependence of integral Betti signatures on factors including the sample size and dimension. This is important for assessment of real-world connectivity matrices, as we show that the standard approach to network construction gives rise to (spurious) spherical geometry, with topology dependent on sample dimensions. Finally, we use the manifolds of constant curvature as comparison models to infer curvature underlying real-world datasets coming from neuroscience, finance and climate. Their associated topological features exhibit a hyperbolic character: the integral Betti signatures associated to these datasets sit in between Euclidean and hyperbolic (of small curvature). The potential confounding ``hyperbologenic effect'' of intrinsic low-rank modular structures is also evaluated through simulations.
{"title":"Integral Betti signature confirms the hyperbolic geometry of brain, climate, and financial networks","authors":"Luigi Caputi, Anna Pidnebesna, Jaroslav Hlinka","doi":"arxiv-2406.15505","DOIUrl":"https://doi.org/arxiv-2406.15505","url":null,"abstract":"This paper extends the possibility to examine the underlying curvature of\u0000data through the lens of topology by using the Betti curves, tools of\u0000Persistent Homology, as key topological descriptors, building on the clique\u0000topology approach. It was previously shown that Betti curves distinguish random\u0000from Euclidean geometric matrices - i.e. distance matrices of points randomly\u0000distributed in a cube with Euclidean distance. In line with previous\u0000experiments, we consider their low-dimensional approximations named integral\u0000Betti values, or signatures that effectively distinguish not only Euclidean,\u0000but also spherical and hyperbolic geometric matrices, both from purely random\u0000matrices as well as among themselves. To prove this, we analyse the behaviour\u0000of Betti curves for various geometric matrices -- i.e. distance matrices of\u0000points randomly distributed on manifolds of constant sectional curvature,\u0000considering the classical models of curvature 0, 1, -1, given by the Euclidean\u0000space, the sphere, and the hyperbolic space. We further investigate the\u0000dependence of integral Betti signatures on factors including the sample size\u0000and dimension. This is important for assessment of real-world connectivity\u0000matrices, as we show that the standard approach to network construction gives\u0000rise to (spurious) spherical geometry, with topology dependent on sample\u0000dimensions. Finally, we use the manifolds of constant curvature as comparison\u0000models to infer curvature underlying real-world datasets coming from\u0000neuroscience, finance and climate. Their associated topological features\u0000exhibit a hyperbolic character: the integral Betti signatures associated to\u0000these datasets sit in between Euclidean and hyperbolic (of small curvature).\u0000The potential confounding ``hyperbologenic effect'' of intrinsic low-rank\u0000modular structures is also evaluated through simulations.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"23 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141520790","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}
Income inequalities and redistribution policies are modeled with a minimal, endogenous model of a simple foraging economy. The model is scaled to match human lifespans and overall death rates. Stochastic income distributions from the model are compared to empirical data from actual economies. Empirical data are fit to implied distributions providing necessary resolution for comparison. The impacts of redistribution policies on total wealth, income distributions, and inequality are shown to be similar for the empirical data and the model. These comparisons enable detailed determinations of population welfare beyond what is possible with total wealth and inequality metrics. Estate taxes in the model appear quite effective in reducing inequality without reducing total wealth. Significant income inequality emerges for the model for a population of equally capable individuals presented with equal opportunities. Stochastic population instability at both the high and low ends of infertility are considered.
{"title":"Death, Taxes, and Inequality. Can a Minimal Model Explain Real Economic Inequality?","authors":"John C. Stevenson","doi":"arxiv-2406.13789","DOIUrl":"https://doi.org/arxiv-2406.13789","url":null,"abstract":"Income inequalities and redistribution policies are modeled with a minimal,\u0000endogenous model of a simple foraging economy. The model is scaled to match\u0000human lifespans and overall death rates. Stochastic income distributions from\u0000the model are compared to empirical data from actual economies. Empirical data\u0000are fit to implied distributions providing necessary resolution for comparison.\u0000The impacts of redistribution policies on total wealth, income distributions,\u0000and inequality are shown to be similar for the empirical data and the model.\u0000These comparisons enable detailed determinations of population welfare beyond\u0000what is possible with total wealth and inequality metrics. Estate taxes in the\u0000model appear quite effective in reducing inequality without reducing total\u0000wealth. Significant income inequality emerges for the model for a population of\u0000equally capable individuals presented with equal opportunities. Stochastic\u0000population instability at both the high and low ends of infertility are\u0000considered.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"33 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141532758","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}
A corporate bond trader in a typical sell side institution such as a bank provides liquidity to the market participants by buying/selling securities and maintaining an inventory. Upon receiving a request for a buy/sell price quote (RFQ), the trader provides a quote by adding a spread over a textit{prevalent market price}. For illiquid bonds, the market price is harder to observe, and traders often resort to available benchmark bond prices (such as MarketAxess, Bloomberg, etc.). In cite{Bergault2023ModelingLI}, the concept of textit{Fair Transfer Price} for an illiquid corporate bond was introduced which is derived from an infinite horizon stochastic optimal control problem (for maximizing the trader's expected P&L, regularized by the quadratic variation). In this paper, we consider the same optimization objective, however, we approach the estimation of an optimal bid-ask spread quoting strategy in a data driven manner and show that it can be learned using Reinforcement Learning. Furthermore, we perform extensive outcome analysis to examine the reasonableness of the trained agent's behavior.
{"title":"Reinforcement Learning for Corporate Bond Trading: A Sell Side Perspective","authors":"Samuel Atkins, Ali Fathi, Sammy Assefa","doi":"arxiv-2406.12983","DOIUrl":"https://doi.org/arxiv-2406.12983","url":null,"abstract":"A corporate bond trader in a typical sell side institution such as a bank\u0000provides liquidity to the market participants by buying/selling securities and\u0000maintaining an inventory. Upon receiving a request for a buy/sell price quote\u0000(RFQ), the trader provides a quote by adding a spread over a textit{prevalent\u0000market price}. For illiquid bonds, the market price is harder to observe, and\u0000traders often resort to available benchmark bond prices (such as MarketAxess,\u0000Bloomberg, etc.). In cite{Bergault2023ModelingLI}, the concept of textit{Fair\u0000Transfer Price} for an illiquid corporate bond was introduced which is derived\u0000from an infinite horizon stochastic optimal control problem (for maximizing the\u0000trader's expected P&L, regularized by the quadratic variation). In this paper,\u0000we consider the same optimization objective, however, we approach the\u0000estimation of an optimal bid-ask spread quoting strategy in a data driven\u0000manner and show that it can be learned using Reinforcement Learning.\u0000Furthermore, we perform extensive outcome analysis to examine the\u0000reasonableness of the trained agent's behavior.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"26 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141501641","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 devise a novel method for implied volatility smoothing based on neural operators. The goal of implied volatility smoothing is to construct a smooth surface that links the collection of prices observed at a specific instant on a given option market. Such price data arises highly dynamically in ever-changing spatial configurations, which poses a major limitation to foundational machine learning approaches using classical neural networks. While large models in language and image processing deliver breakthrough results on vast corpora of raw data, in financial engineering the generalization from big historical datasets has been hindered by the need for considerable data pre-processing. In particular, implied volatility smoothing has remained an instance-by-instance, hands-on process both for neural network-based and traditional parametric strategies. Our general operator deep smoothing approach, instead, directly maps observed data to smoothed surfaces. We adapt the graph neural operator architecture to do so with high accuracy on ten years of raw intraday S&P 500 options data, using a single set of weights. The trained operator adheres to critical no-arbitrage constraints and is robust with respect to subsampling of inputs (occurring in practice in the context of outlier removal). We provide extensive historical benchmarks and showcase the generalization capability of our approach in a comparison with SVI, an industry standard parametrization for implied volatility. The operator deep smoothing approach thus opens up the use of neural networks on large historical datasets in financial engineering.
{"title":"Operator Deep Smoothing for Implied Volatility","authors":"Lukas Gonon, Antoine Jacquier, Ruben Wiedemann","doi":"arxiv-2406.11520","DOIUrl":"https://doi.org/arxiv-2406.11520","url":null,"abstract":"We devise a novel method for implied volatility smoothing based on neural\u0000operators. The goal of implied volatility smoothing is to construct a smooth\u0000surface that links the collection of prices observed at a specific instant on a\u0000given option market. Such price data arises highly dynamically in ever-changing\u0000spatial configurations, which poses a major limitation to foundational machine\u0000learning approaches using classical neural networks. While large models in\u0000language and image processing deliver breakthrough results on vast corpora of\u0000raw data, in financial engineering the generalization from big historical\u0000datasets has been hindered by the need for considerable data pre-processing. In\u0000particular, implied volatility smoothing has remained an instance-by-instance,\u0000hands-on process both for neural network-based and traditional parametric\u0000strategies. Our general operator deep smoothing approach, instead, directly\u0000maps observed data to smoothed surfaces. We adapt the graph neural operator\u0000architecture to do so with high accuracy on ten years of raw intraday S&P 500\u0000options data, using a single set of weights. The trained operator adheres to\u0000critical no-arbitrage constraints and is robust with respect to subsampling of\u0000inputs (occurring in practice in the context of outlier removal). We provide\u0000extensive historical benchmarks and showcase the generalization capability of\u0000our approach in a comparison with SVI, an industry standard parametrization for\u0000implied volatility. The operator deep smoothing approach thus opens up the use\u0000of neural networks on large historical datasets in financial engineering.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"46 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141528880","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}
This paper introduces DeepUnifiedMom, a deep learning framework that enhances portfolio management through a multi-task learning approach and a multi-gate mixture of experts. The essence of DeepUnifiedMom lies in its ability to create unified momentum portfolios that incorporate the dynamics of time series momentum across a spectrum of time frames, a feature often missing in traditional momentum strategies. Our comprehensive backtesting, encompassing diverse asset classes such as equity indexes, fixed income, foreign exchange, and commodities, demonstrates that DeepUnifiedMom consistently outperforms benchmark models, even after factoring in transaction costs. This superior performance underscores DeepUnifiedMom's capability to capture the full spectrum of momentum opportunities within financial markets. The findings highlight DeepUnifiedMom as an effective tool for practitioners looking to exploit the entire range of momentum opportunities. It offers a compelling solution for improving risk-adjusted returns and is a valuable strategy for navigating the complexities of portfolio management.
{"title":"DeepUnifiedMom: Unified Time-series Momentum Portfolio Construction via Multi-Task Learning with Multi-Gate Mixture of Experts","authors":"Joel Ong, Dorien Herremans","doi":"arxiv-2406.08742","DOIUrl":"https://doi.org/arxiv-2406.08742","url":null,"abstract":"This paper introduces DeepUnifiedMom, a deep learning framework that enhances\u0000portfolio management through a multi-task learning approach and a multi-gate\u0000mixture of experts. The essence of DeepUnifiedMom lies in its ability to create\u0000unified momentum portfolios that incorporate the dynamics of time series\u0000momentum across a spectrum of time frames, a feature often missing in\u0000traditional momentum strategies. Our comprehensive backtesting, encompassing\u0000diverse asset classes such as equity indexes, fixed income, foreign exchange,\u0000and commodities, demonstrates that DeepUnifiedMom consistently outperforms\u0000benchmark models, even after factoring in transaction costs. This superior\u0000performance underscores DeepUnifiedMom's capability to capture the full\u0000spectrum of momentum opportunities within financial markets. The findings\u0000highlight DeepUnifiedMom as an effective tool for practitioners looking to\u0000exploit the entire range of momentum opportunities. It offers a compelling\u0000solution for improving risk-adjusted returns and is a valuable strategy for\u0000navigating the complexities of portfolio management.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"20 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141520792","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：