Dynamic Parametric and Nonparametric Hedging: Evidence from the Arab Gulf Equity Markets

Abstract

This paper examines the optimal hedging strategies in the Arab Gulf equity markets using a parametric and a nonparametric dynamic approaches in modeling the conditional variances and covariances of equity returns. The parametric approach is based on a multivariate VAR-GARCH model of daily returns, with BEKK specification of Engle and Kroner (1995), and the nonparametric approach adopts a dynamic system based on Filtered Historical Simulation (FHS) of Barone-Adesi et al. (1999) and nonparametric regression. These approaches are then used to calculate optimal portfolio weights and optimal ratios of hedging long and short positions in the Gulf Cooperation Council major sectors, namely, Service, Financial and Industrial. The results show that the nonparametric approach provides higher hedging effectiveness and hence superior hedging strategies.