Long memory and efficiency of Bitcoin under heavy tails

Liang Wu and Shujuan Chen

Applied Economics, 2020, vol. 52, issue 48, 5298-5309

Abstract: The long memory is usually defined via auto-covariances which further connect with Hurst exponent. The heavy tails in Bitcoin returns can cause infinite auto-covariances which make the analysis of long memory and market efficiency in Bitcoin based on estimation of Hurst exponent inappropriate. Few literatures focus on this problem. We provide two approaches based on shuffling method and rank-ordered technique to this problem, and further combine them to analyse the time-varying efficiency and long memory in Bitcoin using sliding window. Results show that the inefficiency and long memory exist in Bitcoin before 2014 and after mid-2017. Especially, the latest data reveal a recent new change that the Bitcoin market has become inefficient and exhibited long memory behaviour since mid-2017, but is turning back to efficiency recently. This change may be due to the frequent key events of Bitcoin in 2017 and 2018, which can break the weak efficiency of Bitcoin. The heavy negative tails with $$\alpha \lt 2$$α

Date: 2020
References: Add references at CitEc
Citations: View citations in EconPapers (8)

Downloads: (external link)
http://hdl.handle.net/10.1080/00036846.2020.1761942 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:applec:v:52:y:2020:i:48:p:5298-5309

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/RAEC20

DOI: 10.1080/00036846.2020.1761942

Access Statistics for this article

Applied Economics is currently edited by Anita Phillips

More articles in Applied Economics from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().