Optimal pairs trading strategies in a cointegration framework

Statistical arbitrage is based on pairs trading of mean-reverting returns. We used cointegrationapproach and ECM-DCC-GARCH to construct 98 pairs of 152 stocks of 3 currencies. Stocks trading isdone by Contract for Difference, a financial derivative product which facilitates short selling andprovides a leverage up to 25 times. To measure the performance of a leveraged strategy, weintroduced the profit factor which is the annualized return rate per unit risk. And the historical risk ismeasured by maximum drawdown. We compared three main strategies: percentage, standarddeviation of cointegration long term residuals and Bollinger Bands (dynamic standard deviation), withand without double confirmation of short term standard deviation modeled by ECM-DCC-GARCH. Eachof the three main strategies is optimized by two optimizers: absolute profit and profit factor. Theoptimization period goes from 2012-01-01 to 2014-12-31, and validation period is from 2015-01-01 to2016-06-01. Our results showed that the USD Bollinger Bands strategy without double confirmationand optimized by profit factor, outperformed other strategies and provided the highest annualizedreturn rate per unit risk