基于Copula-TGARCH模型的股指期货最佳套期保值比研究

Optimal Hedge Ratios of Copula-based Model with Stock Index Futures

本文基于一种新的Copula-TGARCH模型估计股指期货的最佳套期保值比,根据现货和期货收益率序列不同的尾部相依性,用不同的Copula函数形式(Gumbel,Clayton,Gaussian)拟合两者的相关性,并与其它的动态套期保值模型(ECM-CCC-GARCH和ECM-DVEC-GARCH)比较其套期保值的有效性。通过对香港恒生指数现货和期货的实证分析发现:无论样本期内、外,Copula-TGARCH模型的套期保值效果均优于其它模型,而基于非对称Gumbel Copula的套期保值比最佳。

In this paper, I suggest a new copula-based TGARCH model to estimate the optimal hedge ratio, fit the different copula （Gumbel, Clayton, Gaussian）, estimate the parameters of the models by IFM, and compare its effectiveness with that of other hedging models, including ECM-CCC-GARCH and ECM-DVEC-GARCH models. The empirical results shows that in both the in-sample and out-sample tests, the copula-TGARCH hedging model performed better than all of other models, and the hedge ratio based Gumbel copula is the optimal.