为度量高阶矩风险的时变特征,将高频“已实现”二阶矩扩展到“已实现”高阶矩,给出一元及多元“已实现”高阶矩的计算方法;基于效用函数的Taylor展开解决了动态资产配置问题;将“已实现”高阶矩风险测度应用于动态组合投资分析中,推导出带有高阶矩风险的动态投资组合策略,弥补了传统组合投资理论没有考虑高阶矩风险和静态处理问题两大缺陷。选取中国股市的高频金融时间序列进行实证,结果显示高阶矩风险存在波动聚集性,动态组合投资效果优于静态组合投资效果。
Unitvariate and multivariate realized higher moments, which are the expansion of realized volatility, are proposed in the paper to measure the time- varying fmaneial risk. The dynamic assets allocation is settled by Taylor series expansion of utility function. The authors apply realized higher moments to portfolio analysis, and derive dynamic portfolio strategy. Our model repair two defects in traditional portfolio theory without considering higher moments risk and settle problem statically. High frequency financial time series in Chinese stock markets are selected to make empirical research. The empirical results show that higher moments risk possess volatility cluster, and dynamic portfolio is obviously superior to static portfolio.