Trading Index Volatility by the Pair: A Pairs Trading Strategy Built on Fractionally Cointegrated Realized Volatilities

Abstract

Traditional pairs trading approaches are often confined to equities trading. While there exists numerous methods for selection of trading pairs and strategy implementations, minimal work has been conducted on how to incorporate fractional cointegration in a pairs trading setting. This research attempts to fill this gap by constructing a pairs trading strategy that utilizes the fractionally cointegrated properties of the daily realized volatilities of four major global stock indices: the DAX 30, FTSE 100, NIKKEI 225 and the S&P 500. By trading at-the-money straddles using European options on each index, we create a volatility driven strategy that captures mean-reverting properties present in stock market indices. To refine the strategy, we employ trading criteria that accounts for the discrepancy between realized volatility estimates and the implied volatility inherent in the traded options. We find that, under specific investment rules, the trading strategy can generate an optimal portfolio that performs substantially better against an alternative base case measure. Cumulative returns of 48.39% and 51.02% are generated over two testing periods. The strategy, however, is sensitive to thresholds placed on investment rules but nonetheless maintains most market neutrality characteristics.