Combining Mean Reversion and Momentum Trading Strategies in Foreign Exchange Markets

Direct link to article

This paper demonstrates the success of a series of mean-reversion, momentum and combination trading strategies originally designed for use in equities when applied to foreign exchange markets. Returns are measured in deviations from UIP (uncovered interest parity) which states that the changes in exchange rates should incorporate any interest rate differentials between two currencies. The paper concludes that UIP can be exploited using the hybrid mean-reversion/momentum strategy due to inefficiencies in UIP.

Existing research indicates FX returns show strong positive correlations in the short-run (momentum behavior) and negative correlations in the long-run (mean reverting behavior). This study demonstrates a strategy that is a combination of short-run momentum and long-run mean reversion of the deviations of UIP yields abnormal returns when applied to the FX market. The strategy is applied to interest rate parity deviations through a study of 10 developed and highly-liquid currencies over the period of 1978-2008, with USD as the base currency. Three portfolios are tested: mean-reversion only, momentum only, and combination momentum/mean reversion. Each strategy produces statistically significant results.

The return for each currency is estimated using OLS and a portfolio is formed by estimating the highest expected returns less the lowest expected returns (max – min) and holding positions for the next K months, where K = 1, 3, 6, 9 or 12. The results of the equations show the highest average returns for the combined mean-reversion/momentum strategy. The deviations show the mean returns start out positive and revert after about one year when the momentum effect disappears. However, starting around the fourth year, the mean reversion strategy produces positive returns.

This paper also compares the similarities between the behavior of the FX market and that of the equities market and concludes this resemblance is due to similar behavioral biases operating in both markets, leading to similar inefficiencies. In the stock market, investors often over-react to new information and create momentum, which over time, reverts back to the mean. In the FX market, exchange rates tend to have an overly eager response to monetary policy, however they then revert to equilibrium over the long run. The strategy was originally designed to be used for equity trading however this study shows the strategy performs better in FX markets than in equity markets, and produces a higher Sharpe ratio than common FX trading strategies.