過去的研究指出，固定的避險比率並不能達到最佳的避險效果，而建議最適避險比率應隨時間變動調整。本篇論文提出以Copula-based GARCH模型估計避險比率，並將其避險績效與傳統避險模型 (OLS)，CCC GARCH模型，以及DCC GARCH模型相比較。Copula-based GARCH模型不受限於常態分配的假設，使資產邊際分配的選擇更有彈性，更能貼近其真實的分配。實證結果指出，不論是在樣本內或樣本外，以Copula-based GARCH模型估計避險比率所形成的投資組合變異數皆為最小，有最好的避險績效。 Many recent studies have demonstrated that using the constant hedge ratio obtained by the ordinary least squares method is inappropriate and hence different dynamic hedging strategies are suggested. In this paper we propose a new copula-based GARCH model to estimate the optimal hedge ratio, and compare its hedging effectiveness with different hedge models, including the constant conditional correlation GARCH model and the dynamic conditional correlation GARCH model. The advantage of the proposed model is that it allows for a more flexible distribution specification; Namely, marginal distributions or the dependence structure can be considered separately and simultaneously without the multivariate normality assumption. Hedging performance, in terms of variance reduction of portfolio returns, is evaluated for alternative models. Based on in-sample and out-of-sample comparisons, we find that the proposed model provides best hedging effectiveness.
