Dynamic Hedge Rations on Currency Futures

Abstract

In the globalized economy many businesses are exposed to the foreign exchange risk in their daily trading activities. Exchange traded futures contracts are one of the instruments that are designed specifically to hedge such risk. Over the years researchers and practitioners have been interested in designing the optimal hedge ratio as a number of contracts that should be purchased in order to minimize the variance of the hedged portfolio. Early methods of calculating that ratio assumed time invariant variance covariance structure between spot and futures prices resulting in static ratios. However this assumption has been challenged and models that allow for dynamic evolution of variances and covariances gained on popularity. The purpose of this paper is to investigate the performance of the dynamic hedge ratio strategy on EUR/SEK and USD/SEK designed with bivariate error correction GARCH model with diagonal BEKK parameterization. Also we compare this strategy with other most commonly used hedging schemes such as static OLS and naïve hedging. Using daily observations spanning over almost 8 years we found that despite the theoretical feasibility of the bivariate GARCH it fails to outperform static regression based hedges both in and out of sample.