arXiv - QuantFin - Portfolio Management最新文献_第10页

On optimal tracking portfolio in incomplete markets: The classical control and the reinforcement learning approaches 不完全市场中最优跟踪投资组合:经典控制与强化学习方法

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-24 DOI: arxiv-2311.14318

Lijun Bo, Yijie Huang, Xiang Yu

{"title":"On optimal tracking portfolio in incomplete markets: The classical control and the reinforcement learning approaches","authors":"Lijun Bo, Yijie Huang, Xiang Yu","doi":"arxiv-2311.14318","DOIUrl":"https://doi.org/arxiv-2311.14318","url":null,"abstract":"This paper studies an infinite horizon optimal tracking portfolio problem\u0000using capital injection in incomplete market models. We consider the benchmark\u0000process modelled by a geometric Brownian motion with zero drift driven by some\u0000unhedgeable risk. The relaxed tracking formulation is adopted where the\u0000portfolio value compensated by the injected capital needs to outperform the\u0000benchmark process at any time, and the goal is to minimize the cost of the\u0000discounted total capital injection. In the first part, we solve the stochastic\u0000control problem when the market model is known, for which the equivalent\u0000auxiliary control problem with reflections and the associated HJB equation with\u0000a Neumann boundary condition are studied. In the second part, the market model\u0000is assumed to be unknown, for which we consider the exploratory formulation of\u0000the control problem with entropy regularizer and develop the continuous-time\u0000q-learning algorithm for the stochastic control problem with state reflections.\u0000In an illustrative example, we show the satisfactory performance of the\u0000q-learning algorithm.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"6 11","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138494882","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}

引用次数: 0

Curriculum Learning and Imitation Learning for Model-free Control on Financial Time-series 金融时间序列无模型控制的课程学习与模仿学习

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-22 DOI: arxiv-2311.13326

Woosung Koh, Insu Choi, Yuntae Jang, Gimin Kang, Woo Chang Kim

引用次数: 0

High order universal portfolios 高阶通用投资组合

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-22 DOI: arxiv-2311.13564

Gabriel Turinici

引用次数: 0

Can we hedge carbon risk? A network embedding approach 我们能对冲碳风险吗?一种网络嵌入方法

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-21 DOI: arxiv-2311.12450

Michele Azzone, Maria Chiara Pocelli, Davide Stocco

引用次数: 0

Optimal Portfolio with Ratio Type Periodic Evaluation under Short-Selling Prohibition 禁止卖空条件下比率型周期评估的最优投资组合

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-21 DOI: arxiv-2311.12517

Wenyuan Wang, Kaixin Yan, Xiang Yu

引用次数: 0

DeFi Security: Turning The Weakest Link Into The Strongest Attraction DeFi安全:把最薄弱的环节变成最强的吸引力

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-20 DOI: arxiv-2312.00033

Ravi Kashyap

{"title":"DeFi Security: Turning The Weakest Link Into The Strongest Attraction","authors":"Ravi Kashyap","doi":"arxiv-2312.00033","DOIUrl":"https://doi.org/arxiv-2312.00033","url":null,"abstract":"The primary innovation we pioneer -- focused on blockchain information\u0000security -- is called the Safe-House. The Safe-House is badly needed since\u0000there are many ongoing hacks and security concerns in the DeFi space right now.\u0000The Safe-House is a piece of engineering sophistication that utilizes existing\u0000blockchain principles to bring about greater security when customer assets are\u0000moved around. The Safe-House logic is easily implemented as smart contracts on\u0000any decentralized system. The amount of funds at risk from both internal and\u0000external parties -- and hence the maximum one time loss -- is guaranteed to\u0000stay within the specified limits based on cryptographic fundamentals. To improve the safety of the Safe-House even further, we adapt the one time\u0000password (OPT) concept to operate using blockchain technology. Well suited to\u0000blockchain cryptographic nuances, our secondary advancement can be termed the\u0000one time next time password (OTNTP) mechanism. The OTNTP is designed to\u0000complement the Safe-House making it even more safe. We provide a detailed threat assessment model -- discussing the risks faced\u0000by DeFi protocols and the specific risks that apply to blockchain fund\u0000management -- and give technical arguments regarding how these threats can be\u0000overcome in a robust manner. We discuss how the Safe-House can participate with\u0000other external yield generation protocols in a secure way. We provide reasons\u0000for why the Safe-House increases safety without sacrificing the efficiency of\u0000operation. We start with a high level intuitive description of the landscape,\u0000the corresponding problems and our solutions. We then supplement this overview\u0000with detailed discussions including the corresponding mathematical formulations\u0000and pointers for technological implementation. This approach ensures that the\u0000article is accessible to a broad audience.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"118 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138516967","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}

引用次数: 0

Optimal Retirement Choice under Age-dependent Force of Mortality 年龄依赖性死亡力下的最优退休选择

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-20 DOI: arxiv-2311.12169

Giorgio Ferrari, Shihao Zhu

{"title":"Optimal Retirement Choice under Age-dependent Force of Mortality","authors":"Giorgio Ferrari, Shihao Zhu","doi":"arxiv-2311.12169","DOIUrl":"https://doi.org/arxiv-2311.12169","url":null,"abstract":"This paper examines the retirement decision, optimal investment, and\u0000consumption strategies under an age-dependent force of mortality. We formulate\u0000the optimization problem as a combined stochastic control and optimal stopping\u0000problem with a random time horizon, featuring three state variables: wealth,\u0000labor income, and force of mortality. To address this problem, we transform it\u0000into its dual form, which is a finite time horizon, three-dimensional\u0000degenerate optimal stopping problem with interconnected dynamics. We establish\u0000the existence of an optimal retirement boundary that splits the state space\u0000into continuation and stopping regions. Regularity of the optimal stopping\u0000value function is derived and the boundary is proved to be Lipschitz\u0000continuous, and it is characterized as the unique solution to a nonlinear\u0000integral equation, which we compute numerically. In the original coordinates,\u0000the agent thus retires whenever her wealth exceeds an age-, labor income- and\u0000mortality-dependent transformed version of the optimal stopping boundary. We\u0000also provide numerical illustrations of the optimal strategies, including the\u0000sensitivities of the optimal retirement boundary concerning the relevant\u0000model's parameters.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"7 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138494857","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}

引用次数: 0

Optimal Transport Divergences induced by Scoring Functions 由评分函数引起的最优运输散度

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-20 DOI: arxiv-2311.12183

Silvana M. Pesenti, Steven Vanduffel

引用次数: 0

Sector Rotation by Factor Model and Fundamental Analysis 通过因子模型和基本面分析进行行业轮换

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-18 DOI: arxiv-2401.00001

Runjia Yang, Beining Shi

引用次数: 0

Reinforcement Learning with Maskable Stock Representation for Portfolio Management in Customizable Stock Pools 可自定义股票池投资组合管理的可掩码股票表示强化学习

arXiv - QuantFin - Portfolio Management Pub Date : 2023-11-17 DOI: arxiv-2311.10801

Wentao Zhang, Yilei Zhao, Shuo Sun, Jie Ying, Yonggang Xie, Zitao Song, Xinrun Wang, Bo An

{"title":"Reinforcement Learning with Maskable Stock Representation for Portfolio Management in Customizable Stock Pools","authors":"Wentao Zhang, Yilei Zhao, Shuo Sun, Jie Ying, Yonggang Xie, Zitao Song, Xinrun Wang, Bo An","doi":"arxiv-2311.10801","DOIUrl":"https://doi.org/arxiv-2311.10801","url":null,"abstract":"Portfolio management (PM) is a fundamental financial trading task, which\u0000explores the optimal periodical reallocation of capitals into different stocks\u0000to pursue long-term profits. Reinforcement learning (RL) has recently shown its\u0000potential to train profitable agents for PM through interacting with financial\u0000markets. However, existing work mostly focuses on fixed stock pools, which is\u0000inconsistent with investors' practical demand. Specifically, the target stock\u0000pool of different investors varies dramatically due to their discrepancy on\u0000market states and individual investors may temporally adjust stocks they desire\u0000to trade (e.g., adding one popular stocks), which lead to customizable stock\u0000pools (CSPs). Existing RL methods require to retrain RL agents even with a tiny\u0000change of the stock pool, which leads to high computational cost and unstable\u0000performance. To tackle this challenge, we propose EarnMore, a rEinforcement\u0000leARNing framework with Maskable stOck REpresentation to handle PM with CSPs\u0000through one-shot training in a global stock pool (GSP). Specifically, we first\u0000introduce a mechanism to mask out the representation of the stocks outside the\u0000target pool. Second, we learn meaningful stock representations through a\u0000self-supervised masking and reconstruction process. Third, a re-weighting\u0000mechanism is designed to make the portfolio concentrate on favorable stocks and\u0000neglect the stocks outside the target pool. Through extensive experiments on 8\u0000subset stock pools of the US stock market, we demonstrate that EarnMore\u0000significantly outperforms 14 state-of-the-art baselines in terms of 6 popular\u0000financial metrics with over 40% improvement on profit.","PeriodicalId":501045,"journal":{"name":"arXiv - QuantFin - Portfolio Management","volume":"7 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138494856","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}

引用次数: 0