Hedging short‐term interest risk under time‐varying distributions

Abstract

We investigate the performance of conditional hedging strategies in the context of banker's acceptance positions. This strategy is based on the GARCH methodology developed by Engle (1982) and Bollerslev (1986), and incorporates information contained in past return innovations as well as past conditional variances. We extend previous research byallowing asymmetries in the volatility response to positive shocks and to negative shocks in the construction of our hedged positions, and by relaxing the constant conditional correlation assumption imposed by previous researchers. Our evidence not only supports earlier findings suggesting that the conditional hedging strategy outperforms the constant hedge model, both statistically and economically, but also shows that even greater risk reduction may be achieved by accounting for asymmetries in the spot-futures joint dynamics. Our results also indicate that the constant hedging model is superior to the widely based duration-based approach. This study represents the first investigation of the Montreal Exchange's BA futures contract (BAX).