arXiv - QuantFin - Computational Finance最新文献_第7页

Markowitz Meets Bellman: Knowledge-distilled Reinforcement Learning for Portfolio Management 马科维茨遇上贝尔曼：投资组合管理中的知识强化学习

arXiv - QuantFin - Computational Finance Pub Date : 2024-05-08 DOI: arxiv-2405.05449

Gang Hu, Ming Gu

引用次数: 0

A weighted multilevel Monte Carlo method 加权多级蒙特卡罗方法

arXiv - QuantFin - Computational Finance Pub Date : 2024-05-06 DOI: arxiv-2405.03453

Yu Li, Antony Ware

{"title":"A weighted multilevel Monte Carlo method","authors":"Yu Li, Antony Ware","doi":"arxiv-2405.03453","DOIUrl":"https://doi.org/arxiv-2405.03453","url":null,"abstract":"The Multilevel Monte Carlo (MLMC) method has been applied successfully in a\u0000wide range of settings since its first introduction by Giles (2008). When using\u0000only two levels, the method can be viewed as a kind of control-variate approach\u0000to reduce variance, as earlier proposed by Kebaier (2005). We introduce a\u0000generalization of the MLMC formulation by extending this control variate\u0000approach to any number of levels and deriving a recursive formula for computing\u0000the weights associated with the control variates and the optimal numbers of\u0000samples at the various levels. We also show how the generalisation can also be applied to the\u0000emph{multi-index} MLMC method of Haji-Ali, Nobile, Tempone (2015), at the cost\u0000of solving a $(2^d-1)$-dimensional minimisation problem at each node when $d$\u0000index dimensions are used. The comparative performance of the weighted MLMC method is illustrated in a\u0000range of numerical settings. While the addition of weights does not change the\u0000emph{asymptotic} complexity of the method, the results show that significant\u0000efficiency improvements over the standard MLMC formulation are possible,\u0000particularly when the coarse level approximations are poorly correlated.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140882170","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

Modelling Opaque Bilateral Market Dynamics in Financial Trading: Insights from a Multi-Agent Simulation Study 金融交易中的不透明双边市场动态建模：多代理模拟研究的启示

arXiv - QuantFin - Computational Finance Pub Date : 2024-05-05 DOI: arxiv-2405.02849

Alicia Vidler, Toby Walsh

{"title":"Modelling Opaque Bilateral Market Dynamics in Financial Trading: Insights from a Multi-Agent Simulation Study","authors":"Alicia Vidler, Toby Walsh","doi":"arxiv-2405.02849","DOIUrl":"https://doi.org/arxiv-2405.02849","url":null,"abstract":"Exploring complex adaptive financial trading environments through multi-agent\u0000based simulation methods presents an innovative approach within the realm of\u0000quantitative finance. Despite the dominance of multi-agent reinforcement\u0000learning approaches in financial markets with observable data, there exists a\u0000set of systematically significant financial markets that pose challenges due to\u0000their partial or obscured data availability. We, therefore, devise a\u0000multi-agent simulation approach employing small-scale meta-heuristic methods.\u0000This approach aims to represent the opaque bilateral market for Australian\u0000government bond trading, capturing the bilateral nature of bank-to-bank\u0000trading, also referred to as \"over-the-counter\" (OTC) trading, and commonly\u0000occurring between \"market makers\". The uniqueness of the bilateral market,\u0000characterized by negotiated transactions and a limited number of agents, yields\u0000valuable insights for agent-based modelling and quantitative finance. The\u0000inherent rigidity of this market structure, which is at odds with the global\u0000proliferation of multilateral platforms and the decentralization of finance,\u0000underscores the unique insights offered by our agent-based model. We explore\u0000the implications of market rigidity on market structure and consider the\u0000element of stability, in market design. This extends the ongoing discourse on\u0000complex financial trading environments, providing an enhanced understanding of\u0000their dynamics and implications.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"18 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140882247","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

Gradient-enhanced sparse Hermite polynomial expansions for pricing and hedging high-dimensional American options 用于高维美式期权定价和对冲的梯度增强稀疏赫米特多项式展开法

arXiv - QuantFin - Computational Finance Pub Date : 2024-05-04 DOI: arxiv-2405.02570

Jiefei Yang, Guanglian Li

{"title":"Gradient-enhanced sparse Hermite polynomial expansions for pricing and hedging high-dimensional American options","authors":"Jiefei Yang, Guanglian Li","doi":"arxiv-2405.02570","DOIUrl":"https://doi.org/arxiv-2405.02570","url":null,"abstract":"We propose an efficient and easy-to-implement gradient-enhanced least squares\u0000Monte Carlo method for computing price and Greeks (i.e., derivatives of the\u0000price function) of high-dimensional American options. It employs the sparse\u0000Hermite polynomial expansion as a surrogate model for the continuation value\u0000function, and essentially exploits the fast evaluation of gradients. The\u0000expansion coefficients are computed by solving a linear least squares problem\u0000that is enhanced by gradient information of simulated paths. We analyze the\u0000convergence of the proposed method, and establish an error estimate in terms of\u0000the best approximation error in the weighted $H^1$ space, the statistical error\u0000of solving discrete least squares problems, and the time step size. We present\u0000comprehensive numerical experiments to illustrate the performance of the\u0000proposed method. The results show that it outperforms the state-of-the-art\u0000least squares Monte Carlo method with more accurate price, Greeks, and optimal\u0000exercise strategies in high dimensions but with nearly identical computational\u0000cost, and it can deliver comparable results with recent neural network-based\u0000methods up to dimension 100.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"20 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140882372","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

Fourier-Laplace transforms in polynomial Ornstein-Uhlenbeck volatility models 多项式奥恩斯坦-乌伦贝克波动率模型中的傅立叶-拉普拉斯变换

arXiv - QuantFin - Computational Finance Pub Date : 2024-05-03 DOI: arxiv-2405.02170

Eduardo Abi JaberXiaoyuan, ShaunXiaoyuan, Li, Xuyang Lin

引用次数: 0

DAM: A Universal Dual Attention Mechanism for Multimodal Timeseries Cryptocurrency Trend Forecasting DAM：用于多模态时间序列加密货币趋势预测的通用双重关注机制

arXiv - QuantFin - Computational Finance Pub Date : 2024-05-01 DOI: arxiv-2405.00522

Yihang Fu, Mingyu Zhou, Luyao Zhang

{"title":"DAM: A Universal Dual Attention Mechanism for Multimodal Timeseries Cryptocurrency Trend Forecasting","authors":"Yihang Fu, Mingyu Zhou, Luyao Zhang","doi":"arxiv-2405.00522","DOIUrl":"https://doi.org/arxiv-2405.00522","url":null,"abstract":"In the distributed systems landscape, Blockchain has catalyzed the rise of\u0000cryptocurrencies, merging enhanced security and decentralization with\u0000significant investment opportunities. Despite their potential, current research\u0000on cryptocurrency trend forecasting often falls short by simplistically merging\u0000sentiment data without fully considering the nuanced interplay between\u0000financial market dynamics and external sentiment influences. This paper\u0000presents a novel Dual Attention Mechanism (DAM) for forecasting cryptocurrency\u0000trends using multimodal time-series data. Our approach, which integrates\u0000critical cryptocurrency metrics with sentiment data from news and social media\u0000analyzed through CryptoBERT, addresses the inherent volatility and prediction\u0000challenges in cryptocurrency markets. By combining elements of distributed\u0000systems, natural language processing, and financial forecasting, our method\u0000outperforms conventional models like LSTM and Transformer by up to 20% in\u0000prediction accuracy. This advancement deepens the understanding of distributed\u0000systems and has practical implications in financial markets, benefiting\u0000stakeholders in cryptocurrency and blockchain technologies. Moreover, our\u0000enhanced forecasting approach can significantly support decentralized science\u0000(DeSci) by facilitating strategic planning and the efficient adoption of\u0000blockchain technologies, improving operational efficiency and financial risk\u0000management in the rapidly evolving digital asset domain, thus ensuring optimal\u0000resource allocation.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"24 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140829337","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

The Effect of Data Types' on the Performance of Machine Learning Algorithms for Financial Prediction 数据类型对金融预测机器学习算法性能的影响

arXiv - QuantFin - Computational Finance Pub Date : 2024-04-30 DOI: arxiv-2404.19324

Hulusi Mehmet Tanrikulu, Hakan Pabuccu

{"title":"The Effect of Data Types' on the Performance of Machine Learning Algorithms for Financial Prediction","authors":"Hulusi Mehmet Tanrikulu, Hakan Pabuccu","doi":"arxiv-2404.19324","DOIUrl":"https://doi.org/arxiv-2404.19324","url":null,"abstract":"Forecasting cryptocurrencies as a financial issue is crucial as it provides\u0000investors with possible financial benefits. A small improvement in forecasting\u0000performance can lead to increased profitability; therefore, obtaining a\u0000realistic forecast is very important for investors. Successful forecasting\u0000provides traders with effective buy-or-hold strategies, allowing them to make\u0000more profits. The most important thing in this process is to produce accurate\u0000forecasts suitable for real-life applications. Bitcoin, frequently mentioned\u0000recently due to its volatility and chaotic behavior, has begun to pay great\u0000attention and has become an investment tool, especially during and after the\u0000COVID-19 pandemic. This study provided a comprehensive methodology, including\u0000constructing continuous and trend data using one and seven years periods of\u0000data as inputs and applying machine learning (ML) algorithms to forecast\u0000Bitcoin price movement. A binarization procedure was applied using continuous\u0000data to construct the trend data representing each input feature trend.\u0000Following the related literature, the input features are determined as\u0000technical indicators, google trends, and the number of tweets. Random forest\u0000(RF), K-Nearest neighbor (KNN), Extreme Gradient Boosting (XGBoost-XGB),\u0000Support vector machine (SVM) Naive Bayes (NB), Artificial Neural Networks\u0000(ANN), and Long-Short-Term Memory (LSTM) networks were applied on the selected\u0000features for prediction purposes. This work investigates two main research\u0000questions: i. How does the sample size affect the prediction performance of ML\u0000algorithms? ii. How does the data type affect the prediction performance of ML\u0000algorithms? Accuracy and area under the ROC curve (AUC) values were used to\u0000compare the model performance. A t-test was performed to test the statistical\u0000significance of the prediction results.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"10 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140829346","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

Efficient inverse $Z$-transform and Wiener-Hopf factorization 高效反Z$变换和维纳-霍普夫因式分解

arXiv - QuantFin - Computational Finance Pub Date : 2024-04-30 DOI: arxiv-2404.19290

Svetlana Boyarchenko, Sergei Levendorskiĭ

引用次数: 0

Assessing the Potential of AI for Spatially Sensitive Nature-Related Financial Risks 评估人工智能对空间敏感的自然金融风险的潜力

arXiv - QuantFin - Computational Finance Pub Date : 2024-04-26 DOI: arxiv-2404.17369

Steven Reece, Emma O donnell, Felicia Liu, Joanna Wolstenholme, Frida Arriaga, Giacomo Ascenzi, Richard Pywell

{"title":"Assessing the Potential of AI for Spatially Sensitive Nature-Related Financial Risks","authors":"Steven Reece, Emma O donnell, Felicia Liu, Joanna Wolstenholme, Frida Arriaga, Giacomo Ascenzi, Richard Pywell","doi":"arxiv-2404.17369","DOIUrl":"https://doi.org/arxiv-2404.17369","url":null,"abstract":"There is growing recognition among financial institutions, financial\u0000regulators and policy makers of the importance of addressing nature-related\u0000risks and opportunities. Evaluating and assessing nature-related risks for\u0000financial institutions is challenging due to the large volume of heterogeneous\u0000data available on nature and the complexity of investment value chains and the\u0000various components' relationship to nature. The dual problem of scaling data\u0000analytics and analysing complex systems can be addressed using Artificial\u0000Intelligence (AI). We address issues such as plugging existing data gaps with\u0000discovered data, data estimation under uncertainty, time series analysis and\u0000(near) real-time updates. This report presents potential AI solutions for\u0000models of two distinct use cases, the Brazil Beef Supply Use Case and the Water\u0000Utility Use Case. Our two use cases cover a broad perspective within\u0000sustainable finance. The Brazilian cattle farming use case is an example of\u0000greening finance - integrating nature-related considerations into mainstream\u0000financial decision-making to transition investments away from sectors with poor\u0000historical track records and unsustainable operations. The deployment of\u0000nature-based solutions in the UK water utility use case is an example of\u0000financing green - driving investment to nature-positive outcomes. The two use\u0000cases also cover different sectors, geographies, financial assets and AI\u0000modelling techniques, providing an overview on how AI could be applied to\u0000different challenges relating to nature's integration into finance. This report\u0000is primarily aimed at financial institutions but is also of interest to ESG\u0000data providers, TNFD, systems modellers, and, of course, AI practitioners.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"77 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140810550","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

Subset SSD for enhanced indexation with sector constraints 加强指数化的部门限制的 SSD 子集

arXiv - QuantFin - Computational Finance Pub Date : 2024-04-25 DOI: arxiv-2404.16777

Cristiano Arbex Valle, John E Beasley

{"title":"Subset SSD for enhanced indexation with sector constraints","authors":"Cristiano Arbex Valle, John E Beasley","doi":"arxiv-2404.16777","DOIUrl":"https://doi.org/arxiv-2404.16777","url":null,"abstract":"In this paper we apply second order stochastic dominance (SSD) to the problem\u0000of enhanced indexation with asset subset (sector) constraints. The problem we\u0000consider is how to construct a portfolio that is designed to outperform a given\u0000market index whilst having regard to the proportion of the portfolio invested\u0000in distinct market sectors. In our approach, subset SSD, the portfolio\u0000associated with each sector is treated in a SSD manner. In other words in\u0000subset SSD we actively try to find sector portfolios that SSD dominate their\u0000respective sector indices. However the proportion of the overall portfolio\u0000invested in each sector is not pre-specified, rather it is decided via\u0000optimisation. Computational results are given for our approach as applied to\u0000the S&P~500 over the period $29^{text{th}}$ August 2018 to $29^{text{th}}$\u0000December 2023. This period, over 5 years, includes the Covid pandemic, which\u0000had a significant effect on stock prices. Our results indicate that the scaled\u0000version of our subset SSD approach significantly outperforms the S&P~500 over\u0000the period considered. Our approach also outperforms the standard SSD based\u0000approach to the problem.","PeriodicalId":501294,"journal":{"name":"arXiv - QuantFin - Computational Finance","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140798514","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