Profiting Off the High Correlation of Cryptocurrency Pairs Using Statistical Arbitrage

This paper examines a statistical arbitrage strategy proposed by [1] (Leung T, Nguyen H (2019) Constructing cointegrated cryptocurrency portfolios for statistical arbitrage. Stud Econ Finance 36(4):581–599. ). It aims to exploit correlations between Bitcoin, Ethereum, Litecoin, and Bitcoin Cash. Backtesting on a dataset comprising over 81,000 data points yields a 100% win rate, while live trading implementation sees win rates ranging from 79 to 100%. The live bot generated excess return ranging from −0.4 to 8.1% versus its benchmark. By focusing solely on statistical properties, the strategy offers a macro-agnostic approach to cryptocurrency trading, potentially leading to more stable and predictable outcomes.