The Use of Term Structure Information in the Hedging of Japanese Government Bonds

Abstract

This paper employs the Kalman filter procedure to explore the impact of yield curve factors on the hedging of Japanese government bond (JGB) using treasury futures. Three parameters (i.e., level parameter, slope parameter, and curvature parameter) embedded in Nelson and Siegel (1987) are used to be the proxies of interest rate risk. The out-of-sample hedging performance is also provided by moving window technology. The empirical results indicate that there is a significant relationship between the optimal hedge ratio and the yield curve factors. However, the time varying hedge ratio that includes the yield curve variables from the information set would not provide good out-of-sample hedging effectiveness. But the out-of-sample results also demonstrate that the performance of the time varying hedge ratio with yield curve factors is better than hedge ratio with naive hedge or OLS model.