High frequency multiscale relationships among major cryptocurrencies: portfolio management implications

Walid Mensi; Mobeen Ur Rehman; Muhammad Shafiullah; Khamis Hamed Al-Yahyaee; Ahmet Sensoy

doi:10.1186/s40854-021-00290-w

High frequency multiscale relationships among major cryptocurrencies: portfolio management implications

Financ Innov. 2021;7(1):75. doi: 10.1186/s40854-021-00290-w. Epub 2021 Sep 13.

Authors

Walid Mensi¹, Mobeen Ur Rehman², Muhammad Shafiullah³, Khamis Hamed Al-Yahyaee⁴, Ahmet Sensoy⁵

Affiliations

¹ Department of Economics and Finance, College of Economics and Political Science, Sultan Qaboos University, Muscat, Oman.
² Shaheed Zulfikar Ali Bhutto Institute of Science and Technology (SZABIST), Islamabad, Pakistan.
³ School of Economics, University of Nottingham Malaysia, Jalan Broga, 43500 Semenyih, Selangor Malaysia.
⁴ Muscat University, Al Ghubrah North, Muscat, Oman.
⁵ Faculty of Business Administration, Bilkent University, 06800 Ankara, Turkey.

Abstract

This paper examines the high frequency multiscale relationships and nonlinear multiscale causality between Bitcoin, Ethereum, Monero, Dash, Ripple, and Litecoin. We apply nonlinear Granger causality and rolling window wavelet correlation (RWCC) to 15 min-data. Empirical RWCC results indicate mostly positive co-movements and long-term memory between the cryptocurrencies, especially between Bitcoin, Ethereum, and Monero. The nonlinear Granger causality tests reveal dual causation between most of the cryptocurrency pairs. We advance evidence to improve portfolio risk assessment, and hedging strategies.

Keywords: Cryptocurrency; High frequency analysis; Nonlinear multiscale causality; Rolling window wavelet correlation.