资产收益的跳跃行为给套期保值决策带来了挑战.提出了考虑跳跃、基于预测的VecHAR—RVRCOV—J模型,首次将高频数据中蕴含的跳跃信息引入套期保值决策,对期货和现货收益率的已实现二阶矩做异质滞后阶向量自回归,构造动态套期保值比率的预测统计量.实证应用中以沪深300股指期货及沪深300指数为对象构建套期保值策略,在样本内和样本外的综合套保绩效考核上,新模型优于常用的二元GARCH模型.
Jumps in asset returns bring challenge for hedging decision. This article presents the prediction based VecHAR-RVRCOV-J futures hedging model which allows for jumps. To make the hedging decision, jump variation implied in high-frequency data is integrated for the first time by the new model into the vector heterogeneous autoregressive system for realized second moment of the spot and futures returns. In empirical application, CSI300 futures and its underlying index are used to construct hedging strategy. Both in-sample and out-of-sample performance criteria show that the proposed method is better than conventional bivariate GARCH models.