极端波动、跳跃和尾部风险——基于已实现波动率的股票市场风险动态预测

发布时间：2018-02-20 19:41

本文关键词： 已实现波动率跳跃尾部风险动态预测　出处：《数理统计与管理》2014年01期 　论文类型：期刊论文

【摘要】：本文基于跳跃扩散波动理论,利用非参数方法估计波动中跳跃成份,研究我国股市波动中跳跃的动态演变特征,将跳跃作为股市波动的重要因素纳入模型,建立我国股指收益率的非齐次自回归已实现波动率模型,利用条件极值方法对我国股市的波动风险进行动态预测.统计结果表明:在股市大波动时期跳跃发生更为频繁,而且和连续成份相比较,跳跃对股指波动的贡献占据主要地位。和传统的EGARCH模型相比,包含跳跃的已实现波动率模型对股指波动风险预测性能明显优于EGARCH模型风险预测,跳跃对股指波动风险预测具有显著的解释力.对股市历史数据的分析和预测表明,跳跃对于股指日收益率尾部行为具有重要的影响,大跳跃的发生导致尾部风险显著增大。本文研究结果对于政府监管部门监控股市和制定有效的调控措施加强股市风险管理具有重要的参考价值。
[Abstract]:In this paper, based on the theory of jump diffusion fluctuation, we use nonparametric method to estimate the jump component in volatility, and study the dynamic evolution characteristics of jump in stock market volatility in our country, and put jump as an important factor in stock market volatility into the model. A non-homogeneous autoregressive volatility model of stock index returns in China is established, and the conditional extremum method is used to dynamically predict the volatility risk of the stock market in China. The statistical results show that the jump occurs more frequently in the period of large volatility in the stock market. Compared with the continuous component, the contribution of jump to stock index volatility is more important. Compared with the traditional EGARCH model, the performance of realized volatility model with jump is better than that of EGARCH model. The analysis and prediction of the historical data of stock market show that the jump has an important influence on the tail behavior of the daily yield of stock index. The research results of this paper have important reference value for the government supervision department to monitor the stock market and to formulate effective control measures to strengthen the stock market risk management.
【作者单位】：北京国家会计学院;西安交通大学经济与金融学院;中国人民大学统计学院;
【基金】：中国博士后科学基金(20100471621)
【分类号】：F830.91;F224

【参考文献】