Performance of Pairs Trading on the S&P 500: Distance and Mixed Copula Models

Abstract

We propose an alternative investment strategy for pairs trading using Archimedean copulas in order to cover a wider range of tail dependence patterns and apply it to the S&P 500 stocks from 1990 to 2015. Empirical results show that our mixed copula approach generates higher average and risk adjusted excess returns and lower drawdown risk than the traditional distance method under different weighting structures when we control for trading costs. Particularly, the mixed copula and distance methods show a mean annualized value-weighted excess returns after costs on committed capital of 3.68% and 2.30% for top 5 pairs, with annual Sharpe ratios of 0.58 and 0.28, respectively. In addition, the mixed copula method shows a higher probability of yielding positive returns than the distance approach in different scenarios. The high returns can only partially be explained by common sources of systematic risk. The proposed mixture copulas have been found to be a superior fitting model over different market states, especially during subperiods of a stronger joint tail dependence.