高频数据下商品期货波动率研究——基于不同已实现测度和改进的厚尾分布  

Empirical Study on the Fluctuation of Commodity Futures with High Frequency Data——Based on Different Realized Measures and Error Distributions

作　　者：邓亚东 王波[1] DENG Ya-dong;WANG Bo(School of Management,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学管理学院,上海200093

出　　处：《技术与创新管理》2018年第4期415-420,共6页Technology and Innovation Management

摘　　要：商品期货收益率序列存在明显的厚尾现象,将扰动项设定为厚尾分布的波动率模型要优于普通Gaussian分布以及t分布模型。以7种损失函数作为评价准则,在扰动项分别服从偏t分布以及广义双曲线偏t分布时比较了波动率RV、二次幂变差BV以及medRV3种不同已实现测度下Realized GARCH模型对商品期货的波动性预测能力。最后得到扰动项服从ghst分布的Realized GARCH模型具有更优的预测能力。There is obvious fat tail phenomenon in the yield series of commodity futures,and the volatility model with the fat tailed distribution is better than the ordinary Gaussian distribution and the T distribution model.This paper,using seven kinds of loss function as the evaluation criterion,compared the predictive ability of three realized measures of realized variance,bipower variation and medRV based on skewed T distribution and ghst distribution.Finally,under the three different realized measures,the Realized GARCH model with disturbance subject to the ghst distribution has better prediction ability than skewed T distribution.

关 键 词：Realized GARCH模型 medRV已实现测度 广义双曲线偏t分布 波动率预测 

分 类 号：F830[经济管理—金融学]