Testing for mean reversion in Bitcoin returns with Gibbs-sampling-augmented randomization

Turatti, Douglas Eduardo; Mendes, Fernando Henrique de Paula e Silva; Caldeira, João Frois

doi:10.1016/j.frl.2019.07.025

Cited by 6 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although cryptocurrencies have attracted special attention from academics and practitioners, there lacks a comprehensive study towards the intraday dynamics of cryptocurrencies, and worse, these research have almost exclusively focused on Bitcoin. Among these, Urquhart (2016), Nadarajah and Chu (2017), Bariviera (2017), Yonghong et al (2018),Vidal-Tomas and Ibanez (2018), Turattio et al (2019) and Gerritsen et al (2019) investigate whether time varying behaviour of Bitcoin is predictable, which would be inconsistent with the Efficient Market Hypothesis (EMH) proposing that prices should follow a random walk (Fama, 1970). The same question has been analyzed at the intraday level by Sensoy (2019), Corbet et al (2019) and Akyildirim et al (2019).…”

Section: Introductionmentioning

confidence: 99%

Intraday efficiency-frequency nexus in the cryptocurrency markets

Aslan

Şensoy

2020

Finance Research Letters

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Intraday efficiency-frequency nexus in the cryptocurrency markets

Aslan

Şensoy

2020

Finance Research Letters

View full text Add to dashboard Cite

“…On the one hand, for major coins such as Bitcoin, price movements are subjected to nonlinear temporal dependency (Sensoy, 2019) and long-range dependence (Urquhart, 2016;Bariviera, 2017;Alvarez-Ramirez et al, 2018). The serial dependence is unstable as cryptocurrency returns can exhibit either momentum or mean reversion (Turattia et al, 2020), depending on time horizons and market regimes. Further, the extent of serial dependence varies over time (Bariviera, 2017;Alvarez-Ramirez et al, 2018), echoing the adaptive market hypothesis (Chu et al, 2019) that market efficiency varies over time.…”

Section: Is the Cryptocurrency Market Efficient?mentioning

confidence: 99%

Section: Stream Two: Behaviour Of the Cryptocurrency Marketmentioning

confidence: 99%

Cryptocurrency puzzles: a comprehensive review and re-introduction

Cai,

Xue,

Zhou

2023

JAL

View full text Add to dashboard Cite

Purpose This study reviews existing cryptocurrency research to provide answers to three puzzles in the literature. First, is cryptocurrency more like gold (i.e., a commodity) or should it be classified as a new financial asset? Second, can we apply our knowledge of the traditional capital market to the emerging cryptocurrency market? Third, what might be the future of cryptocurrency?Design/methodology/approach Bibliometric analysis is used to assess 2,098 finance-related cryptocurrency publications from the Web of Science (WoS) Core Collection database from January 2009 to April 2022. Three key research streams are identified, namely, (1) cryptocurrency features, (2) behaviour of the cryptocurrency market and (3) blockchain implications.Findings First, cryptocurrency should be viewed and regulated as a new asset class rather than a currency or a new commodity. While it can provide diversification benefits to the portfolio, cryptocurrency cannot work as a safe haven asset. Second, crypto markets are typically inefficient. Asset bubbles exist and are exacerbated by behavioural finance factors. Third, cryptocurrency demonstrates increasing potential as a medium of exchange and store of value.Originality/value Extant review papers primarily study one or two particular research topics, overlooking the interaction between topics. The few existing systematic literature reviews in this area typically have a narrow focus on trend identification. This study is the first study to provide a comprehensive review of all financial-related studies on cryptocurrency, synthesising the research findings from 2,098 publications to answer three cryptocurrency puzzles.

show abstract

“…Efficiency remains a challenge especially for less liquid cryptocurrencies, and liquidity influences both volatility and return predictability [52], [53], [54], [55]. Explicit mean-reversion in cryptocurrencies in the sense of Jegadeesh [17] has mixed evidence [56], [57], [58].…”

Section: Literature Reviewmentioning

confidence: 99%

Pairs Trading in Cryptocurrency Markets

Fil

Krištoufek

2020

IEEE Access

View full text Add to dashboard Cite

Pairs trading is a strategy based on exploiting mean reversion in prices of securities. Even though these strategies have been shown to perform well for equities, their performance is unknown for the field of cryptocurrencies, usually perceived as inefficient and predictable. We apply the distance and cointegration methods to a basket of 26 liquid cryptocurrencies traded on the Binance exchange, specifically at 5-minute, 1-hour and daily frequencies. In our backtests, the strategies underperform classical benchmarks. However, the results are quite sensitive to parameter settings and external factors such as transaction costs or execution windows. Higher-frequency trading delivers significantly better performance, and while the most common daily distance method returns-0.07% monthly, this increases to 11.61% monthly for 5-minute frequency. Additionally, we find evidence of simple mean-reverting behavior in intraday prices that is missing in daily data, and which provides further support for the inefficiency of cryptocurrency markets. INDEX TERMS cryptocurrencies, intraday trading, pairs trading

show abstract

Testing for mean reversion in Bitcoin returns with Gibbs-sampling-augmented randomization

Cited by 6 publications

References 21 publications

Intraday efficiency-frequency nexus in the cryptocurrency markets

Intraday efficiency-frequency nexus in the cryptocurrency markets

Cryptocurrency puzzles: a comprehensive review and re-introduction

Pairs Trading in Cryptocurrency Markets

Contact Info

Product

Resources

About