加密货币交易量的长记忆和结构性突破

IF 2.5 Q2 ECONOMICS

Eurasian Economic Review Pub Date : 2023-10-16 DOI:10.1007/s40822-023-00238-8

Mohamed Shaker Ahmed, Elie Bouri

{"title":"加密货币交易量的长记忆和结构性突破","authors":"Mohamed Shaker Ahmed, Elie Bouri","doi":"10.1007/s40822-023-00238-8","DOIUrl":null,"url":null,"abstract":"The paper investigates long memory, structural breaks, and spurious long memory in the daily trading volume of the largest and most active cryptocurrencies and stablecoins, namely, Bitcoin, Ethereum, Tether, USD coin, Binance coin, Binance USD, Ripple, Cardano, Solana, Dogecoin and Bitcoin cash. The overall results show that both long memory and structural breaks are present in the cryptocurrencies trading volume, and the detected long memory property is not driven by structural breaks but rather true and thus not spurious. Given this, we conduct out-of-sample forecasting and indicate that the ARFIMA model, which accounts for long-range dependence, has a superior forecasting performance over the standard ARIMA model for four cryptocurrencies, namely, Binance coin, Ripple, Cardano, and Dogecoin at most forecasting horizons ahead and the shorter forecasting horizon (1-day ahead) for most cryptocurrencies under investigation.","PeriodicalId":45064,"journal":{"name":"Eurasian Economic Review","volume":"1 1","pages":"0"},"PeriodicalIF":2.5000,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Long memory and structural breaks of cryptocurrencies trading volume\",\"authors\":\"Mohamed Shaker Ahmed, Elie Bouri\",\"doi\":\"10.1007/s40822-023-00238-8\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The paper investigates long memory, structural breaks, and spurious long memory in the daily trading volume of the largest and most active cryptocurrencies and stablecoins, namely, Bitcoin, Ethereum, Tether, USD coin, Binance coin, Binance USD, Ripple, Cardano, Solana, Dogecoin and Bitcoin cash. The overall results show that both long memory and structural breaks are present in the cryptocurrencies trading volume, and the detected long memory property is not driven by structural breaks but rather true and thus not spurious. Given this, we conduct out-of-sample forecasting and indicate that the ARFIMA model, which accounts for long-range dependence, has a superior forecasting performance over the standard ARIMA model for four cryptocurrencies, namely, Binance coin, Ripple, Cardano, and Dogecoin at most forecasting horizons ahead and the shorter forecasting horizon (1-day ahead) for most cryptocurrencies under investigation.\",\"PeriodicalId\":45064,\"journal\":{\"name\":\"Eurasian Economic Review\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":2.5000,\"publicationDate\":\"2023-10-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Eurasian Economic Review\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1007/s40822-023-00238-8\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ECONOMICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Eurasian Economic Review","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1007/s40822-023-00238-8","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ECONOMICS","Score":null,"Total":0}

引用次数: 0

摘要

本文研究了比特币、以太坊、Tether、USD币、Binance币、Binance USD币、Ripple、Cardano、Solana、Dogecoin和比特币现金等最大、最活跃的加密货币和稳定币的日交易量中的长记忆、结构性断裂和伪长记忆。总体结果表明，加密货币交易量中存在长记忆和结构性中断，并且检测到的长记忆属性不是由结构中断驱动的，而是真实的，因此不是虚假的。鉴于此，我们进行了样本外预测，并表明考虑长期依赖性的ARFIMA模型在大多数预测范围内(即币安币、瑞波币、卡尔达诺币和狗狗币)和大多数正在调查的加密货币的较短预测范围(提前1天)比标准ARIMA模型具有更好的预测性能。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Long memory and structural breaks of cryptocurrencies trading volume

The paper investigates long memory, structural breaks, and spurious long memory in the daily trading volume of the largest and most active cryptocurrencies and stablecoins, namely, Bitcoin, Ethereum, Tether, USD coin, Binance coin, Binance USD, Ripple, Cardano, Solana, Dogecoin and Bitcoin cash. The overall results show that both long memory and structural breaks are present in the cryptocurrencies trading volume, and the detected long memory property is not driven by structural breaks but rather true and thus not spurious. Given this, we conduct out-of-sample forecasting and indicate that the ARFIMA model, which accounts for long-range dependence, has a superior forecasting performance over the standard ARIMA model for four cryptocurrencies, namely, Binance coin, Ripple, Cardano, and Dogecoin at most forecasting horizons ahead and the shorter forecasting horizon (1-day ahead) for most cryptocurrencies under investigation.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Eurasian Economic Review ECONOMICS-

CiteScore

6.00

自引率

2.90%

发文量

期刊介绍： The mission of Eurasian Economic Review is to publish peer-reviewed empirical research papers that test, extend, or build theory and contribute to practice. All empirical methods - including, but not limited to, qualitative, quantitative, field, laboratory, and any combination of methods - are welcome. Empirical, theoretical and methodological articles from all fields of finance and applied macroeconomics are featured in the journal. Theoretical and/or review articles that integrate existing bodies of research and that provide new insights into the field are highly encouraged. The journal has a broad scope, addressing such issues as: financial systems and regulation, corporate and start-up finance, macro and sustainable finance, finance and innovation, consumer finance, public policies on financial markets within local, regional, national and international contexts, money and banking, and the interface of labor and financial economics. The macroeconomics coverage includes topics from monetary economics, labor economics, international economics and development economics. Eurasian Economic Review is published quarterly. To be published in Eurasian Economic Review, a manuscript must make strong empirical and/or theoretical contributions and highlight the significance of those contributions to our field. Consequently, preference is given to submissions that test, extend, or build strong theoretical frameworks while empirically examining issues with high importance for theory and practice. Eurasian Economic Review is not tied to any national context. Although it focuses on Europe and Asia, all papers from related fields on any region or country are highly encouraged. Single country studies, cross-country or regional studies can be submitted.