Detecting Data-Driven Robust Statistical Arbitrage Strategies with Deep Neural Networks

IF 1.4 4区经济学 Q3 BUSINESS, FINANCE

SIAM Journal on Financial Mathematics Pub Date : 2024-05-30 DOI:10.1137/22m1487928

Ariel Neufeld, Julian Sester, Daiying Yin

{"title":"Detecting Data-Driven Robust Statistical Arbitrage Strategies with Deep Neural Networks","authors":"Ariel Neufeld, Julian Sester, Daiying Yin","doi":"10.1137/22m1487928","DOIUrl":null,"url":null,"abstract":"SIAM Journal on Financial Mathematics, Volume 15, Issue 2, Page 436-472, June 2024. <br/> Abstract.We present an approach, based on deep neural networks, for identifying robust statistical arbitrage strategies in financial markets. Robust statistical arbitrage strategies refer to trading strategies that enable profitable trading under model ambiguity. The presented novel methodology allows one to consider a large amount of underlying securities simultaneously and does not depend on the identification of cointegrated pairs of assets; hence it is applicable on high-dimensional financial markets or in markets where classical pairs trading approaches fail. Moreover, we provide a method to build an ambiguity set of admissible probability measures that can be derived from observed market data. Thus, the approach can be considered as being model free and entirely data driven. We showcase the applicability of our method by providing empirical investigations with highly profitable trading performances even in 50 dimensions, during financial crises, and when the cointegration relationship between asset pairs stops to persist.","PeriodicalId":48880,"journal":{"name":"SIAM Journal on Financial Mathematics","volume":"25 1","pages":""},"PeriodicalIF":1.4000,"publicationDate":"2024-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"SIAM Journal on Financial Mathematics","FirstCategoryId":"96","ListUrlMain":"https://doi.org/10.1137/22m1487928","RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"BUSINESS, FINANCE","Score":null,"Total":0}

引用次数: 0

Abstract

SIAM Journal on Financial Mathematics, Volume 15, Issue 2, Page 436-472, June 2024.
Abstract.We present an approach, based on deep neural networks, for identifying robust statistical arbitrage strategies in financial markets. Robust statistical arbitrage strategies refer to trading strategies that enable profitable trading under model ambiguity. The presented novel methodology allows one to consider a large amount of underlying securities simultaneously and does not depend on the identification of cointegrated pairs of assets; hence it is applicable on high-dimensional financial markets or in markets where classical pairs trading approaches fail. Moreover, we provide a method to build an ambiguity set of admissible probability measures that can be derived from observed market data. Thus, the approach can be considered as being model free and entirely data driven. We showcase the applicability of our method by providing empirical investigations with highly profitable trading performances even in 50 dimensions, during financial crises, and when the cointegration relationship between asset pairs stops to persist.

查看原文本刊更多论文

利用深度神经网络检测数据驱动的稳健统计套利策略

SIAM 金融数学期刊》，第 15 卷，第 2 期，第 436-472 页，2024 年 6 月。摘要：我们提出了一种基于深度神经网络的方法，用于识别金融市场中的稳健统计套利策略。稳健统计套利策略指的是在模型模糊的情况下能够实现盈利的交易策略。所提出的新方法可以同时考虑大量的相关证券，并且不依赖于对资产的协整对的识别；因此它适用于高维金融市场或经典对交易方法失效的市场。此外，我们还提供了一种方法，用于建立可从观察到的市场数据中推导出的可接受概率度量的模糊集。因此，我们可以认为这种方法不需要模型，完全由数据驱动。我们通过实证研究展示了我们方法的适用性，即使在 50 维度、金融危机期间以及资产对之间的协整关系不再持续的情况下，我们也能实现高盈利的交易表现。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

SIAM Journal on Financial Mathematics MATHEMATICS, INTERDISCIPLINARY APPLICATIONS-

CiteScore

2.30

自引率

10.00%

发文量

期刊介绍： SIAM Journal on Financial Mathematics (SIFIN) addresses theoretical developments in financial mathematics as well as breakthroughs in the computational challenges they encompass. The journal provides a common platform for scholars interested in the mathematical theory of finance as well as practitioners interested in rigorous treatments of the scientific computational issues related to implementation. On the theoretical side, the journal publishes articles with demonstrable mathematical developments motivated by models of modern finance. On the computational side, it publishes articles introducing new methods and algorithms representing significant (as opposed to incremental) improvements on the existing state of affairs of modern numerical implementations of applied financial mathematics.