Pairs trading in crypto currencies : a cointegration based application

• Main Author: Setzer, Sebastian Heinz
• Publication Date: 2022
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10400.14/40628
• topic: Pairs trading; Crypto currencies; Cointegration

This thesis shows the possibilities of statistical arbitrage in the crypto currency market. Within the field of statistical arbitrage a pairs trading approach is chosen and an overview of deferent pairs trading strategies is given. Finally the cointegration approach in the Engle-Granger two step framework is applied. The approach is clustered into a formation period of 210 days and a trading period of 75 days. Although, it can be shown that the pairs trading outperforms a simple Buy-and-Hold strategy for three independently chosen periods, the returns can not be considered as market neutral in most cases. Furthermore, two strategies in the cointegration setting are compared. One strategy based on daily data and one strategy based on hourly data. It can be shown that the strategy based on daily data is superior to the strategy based on hourly data. This suggests that the crpyto market is not as inefficient as assumed for higher frequency data after all and contradicts results from previous research. Following these results, it can be assumed that the crypto market is still partially not fully efficient.