Intraday Risk Management of Cryptocurrency Returns During 2020–2021 Upsurge: A Conditional EVT Approach

Q2 Business, Management and Accounting

Business Perspectives and Research Pub Date : 2023-03-10 DOI:10.1177/22785337221148878

A. Roy

{"title":"Intraday Risk Management of Cryptocurrency Returns During 2020–2021 Upsurge: A Conditional EVT Approach","authors":"A. Roy","doi":"10.1177/22785337221148878","DOIUrl":null,"url":null,"abstract":"The cryptocurrency market is characterized by extremely high volatility. In the present study, we show the predictive ability of conditional EVT models in the cryptocurrency market during the price upsurge of 2020–2021. Taking high-frequency intraday data of four popular cryptocurrencies, Bitcoin, Ethereum, Litecoin, and Binance coin, we compare the accuracy of different competing models in estimating intraday value at risk (VaR) and expected shortfall (ES). The present study focuses on the extreme value theory (EVT) for modeling the tail of the distribution to forecast the measures of intraday VaR and ES. The study confirms the fat-tailed behavior of intraday returns of all four cryptocurrencies. Further, the study shows the magnitudes of high negative shocks are more than the positive ones for the returns of all four cryptocurrencies. The study uses suitable GARCH-family models such as apARCH, EGARCH, and CGARCH in the ARMA-GARCH framework. Using a two-stage approach the study shows how GARCH-EVT models with skewed student’s— t distribution outperform the predictability of conditional EVT with standard normal distribution as well as the unconditional EVT models in predicting intraday VaR and ES. The result of the study is useful for risk managers, day traders, and also for machine-based algorithmic trading.","PeriodicalId":37330,"journal":{"name":"Business Perspectives and Research","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Business Perspectives and Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1177/22785337221148878","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Business, Management and Accounting","Score":null,"Total":0}

引用次数: 0

Abstract

The cryptocurrency market is characterized by extremely high volatility. In the present study, we show the predictive ability of conditional EVT models in the cryptocurrency market during the price upsurge of 2020–2021. Taking high-frequency intraday data of four popular cryptocurrencies, Bitcoin, Ethereum, Litecoin, and Binance coin, we compare the accuracy of different competing models in estimating intraday value at risk (VaR) and expected shortfall (ES). The present study focuses on the extreme value theory (EVT) for modeling the tail of the distribution to forecast the measures of intraday VaR and ES. The study confirms the fat-tailed behavior of intraday returns of all four cryptocurrencies. Further, the study shows the magnitudes of high negative shocks are more than the positive ones for the returns of all four cryptocurrencies. The study uses suitable GARCH-family models such as apARCH, EGARCH, and CGARCH in the ARMA-GARCH framework. Using a two-stage approach the study shows how GARCH-EVT models with skewed student’s— t distribution outperform the predictability of conditional EVT with standard normal distribution as well as the unconditional EVT models in predicting intraday VaR and ES. The result of the study is useful for risk managers, day traders, and also for machine-based algorithmic trading.

查看原文本刊更多论文

2020-2021年热潮期间加密货币回报的盘中风险管理:一种有条件的EVT方法

加密货币市场的特点是波动性极高。在本研究中，我们展示了条件EVT模型在2020-2021年价格飙升期间在加密货币市场中的预测能力。以四种流行的加密货币比特币、以太坊、莱特币和币安币的高频日内数据为例，我们比较了不同竞争模型在估计日内风险值(VaR)和预期缺口(ES)方面的准确性。本文主要研究利用极值理论(EVT)对分布尾部进行建模，以预测日内VaR和ES的度量。该研究证实了所有四种加密货币的日内回报的肥尾行为。此外，研究表明，对所有四种加密货币的回报来说，高负冲击的幅度大于正冲击的幅度。本研究在ARMA-GARCH框架中使用了合适的garch家族模型，如apARCH、EGARCH和CGARCH。使用两阶段的方法，研究显示了GARCH-EVT模型在预测日内VaR和ES方面如何优于标准正态分布的条件EVT模型以及无条件EVT模型。这项研究的结果对风险经理、日内交易者以及基于机器的算法交易都很有用。

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

Business Perspectives and Research Business, Management and Accounting-Business and International Management

CiteScore

5.00

自引率

0.00%

发文量

期刊介绍： Business Perspectives and Research (BPR) aims to publish conceptual, empirical and applied research. The empirical research published in BPR focuses on testing, extending and building management theory. The goal is to expand and enhance the understanding of business and management through empirical investigation and theoretical analysis. BPR is also a platform for insightful and theoretically strong conceptual and review papers which would contribute to the body of knowledge. BPR seeks to advance the understanding of for-profit and not-for-profit organizations through empirical and conceptual work. It also publishes critical review of newly released books under Book Review section. The aim is to popularize and encourage discussion on ideas expressed in newly released books connected to management and allied disciplines. BPR also periodically publishes management cases grounded in theory, and communications in the form of research notes or comments from researchers and practitioners on published papers for critiquing and/or extending thinking on the area under consideration. The overarching aim of Business Perspectives and Research is to encourage original/innovative thinking through a scientific approach.