An Empirical Analysis of Dynamic Multiscale Hedging using Wavelet Decomposition

摘要

This paper investigates the hedging effectiveness of a dynamic moving windowOLS hedging model, formed using wavelet decomposed time-series. The wavelettransform is applied to calculate the appropriate dynamic minimum-variancehedge ratio for various hedging horizons for a number of assets. Theeffectiveness of the dynamic multiscale hedging strategy is then tested, bothin- and out-of-sample, using standard variance reduction and expanded toinclude a downside risk metric, the time horizon dependent Value-at-Risk.Measured using variance reduction, the effectiveness converges to one at longerscales, while a measure of VaR reduction indicates a portion of residual riskremains at all scales. Analysis of the hedge portfolio distributions indicatethat this unhedged tail risk is related to excess portfolio kurtosis found atall scales.