Risk Minimization and Trading Performance of Dynamic Hedging Models : Time Varying Covariance and Volatility Transmission Effects

The analysis undertaken in this research is a first attempt to comprehensively model all four Samp;P500 markets simultaneously. Synchronously sampled half-hourly observations are generated from transaction data for these four financial assets. Special classes of Simultaneous Volatility (SVL) structures and GARCH models of the variance/covariance matrix and variants augmented with parallel market volatility effects are compared. These are dynamically estimated to minimize out of sample portfolio risk and to generate out of sample hedge ratios for the evaluations. The augmented SVL dominates competing models in terms of trading profits and the out performance is substantial