Cryptocurrencies and Long-Range Trends

Abstract

Abstract In this study we investigate possible long-range trends in the cryptocurrency markets. Our sample includes 37 of the most important cryptocurrencies that reflect more than 80% of the relevant market. For the analysis in the empirical part, we employed the Hurst exponent, a statistical tool used to identify long range autocorrelations and memory in time series data. Our sample covers the period from January 1, 2016 to March 26, 2021. We use three non-overlapping windows for the estimation of the Hurst exponent. With these windows, we assess the dynamic evolution in the structure of the cryptocurrencies market and evaluate the move towards an efficient market. The innovation of this research is that we employ the Hurst exponent that is seldomly used in analyzing this market. Furthermore, the use of both the R/S and DFA analysis and the use of non-overlapping windows enhance our research’s novelty. Finally, we estimate the Hurst for a wide sample of cryptocurrencies that covers more than four fifths of the entire market for the last six years.