A new perspective of the day-of-the-week effect on Bitcoin returns: evidence from an event study hourly approach

Abstract

Research background: A current strand of the financial literature is focusing on detecting inefficiencies, such as the day-of-the-week effect, in the cryptocurrency market. However, these studies are not considering that there are no daily closes in this market, and it is possible to trade cryptocurrencies on a continuous basis. This fact may have led to biases in previous empirical results. Purpose of the article: We propose to analyse the day-of-the-week effect on the Bitcoin from an alternative perspective where each hourly data in a day is considered an event. Focusing on that objective, we employ hourly closing prices for Bitcoin which are taken from the Kraken exchange, one of the world leading exchanges and trading platforms in the cryptocurrency markets, for the period spanning from January 2016 to December 2021. Methods: Contrary to the previous empirical evidence, we do not calculate daily returns, but rather the first stage of our proposed approach is devoted to analysing the hourly mean returns for each of the 24 hours of the day for each day of the week. We look for statistically significant hourly mean returns that could advance the importance of the hourly differentiation in the Bitcoin market. In a second stage, we calculate different post-event cumulative returns which are defined as the change in log prices over a time interval. Finally, we propose different investment strategies simply based on the significant hourly mean returns we obtain and we evaluate their performance in terms of the Sharpe ratio. Findings & value added: We contribute to the debate about the degree of Bitcoin's market efficiency by providing an alternative methodology based on an event study hourly approach. Furthermore, we provide evidence that by investing in different post-event hourly windows it is possible to outperform the classic buy-and-hold strategy.