金融资产收益率波动预测——基于我国股票市场跳跃行为研究

发布时间：2018-01-14 19:04

本文关键词：金融资产收益率波动预测——基于我国股票市场跳跃行为研究　出处：《当代经济科学》2011年05期 　论文类型：期刊论文

【摘要】：金融资产收益率波动是资产定价和金融风险管理的核心部分,而跳跃是收益率波动中的重要组成部分。基于修正Z-检验,本文检测识别我国股市波动中跳跃行为,并且研究了跳跃的时序特征,统计结果表明,在市场大波动时期,和连续成份相比,跳跃对于波动率具有极其重要的贡献。建立包含跳跃的已实现波动率非齐次自回归模型,在波动模型中纳入滞后绝对日收益率和杠杆效应预测股指收益率波动。实证分析结果显示,对于短期的波动预测,包含跳跃和两种影响因素的波动模型表现最好,然而对于提前1月的长期预测,跳跃和连续波动成份分离模型预测明显优于其它模型,这些事实说明跳跃对股指波动率预测具有重要的影响,好坏消息对波动率非对称性具有短期显著影响,而对长期水平的波动率预测影响不显著。
[Abstract]:Volatility of return on financial assets is the core part of asset pricing and financial risk management, and jump is an important part of volatility of return. In this paper, we detect and identify the jumping behavior in Chinese stock market volatility, and study the temporal characteristics of the jump. The statistical results show that in the large volatility period, it is compared with the continuous component. Jump has a very important contribution to volatility. A non-homogeneous autoregressive model of realized volatility including jump is established. In the volatility model, the lag absolute daily rate of return and the leverage effect are included to predict the volatility of stock index. The empirical results show that the short-term volatility is predicted. The volatility model with jump and two influencing factors is the best. However, for the long term prediction of early January, the prediction of jump and continuous fluctuation component separation model is obviously superior to other models. These facts show that the jump has an important influence on the volatility prediction of the stock index, the bad news has a significant impact on the volatility asymmetry in the short term, but the impact on the volatility prediction of the long-term level is not significant.
【作者单位】：西安交通大学经济与金融学院;西北工业大学管理学院;
【基金】：中国博士后科学基金(20100471621) 教育部人文社会科学研究青年基金项目(09XJC790011)
【分类号】：F224;F832.51
【正文快照】： 一、引言金融资产收益率波动是资产定价和金融风险管理的核心部分。金融实证研究表明金融资产收益率的波动呈现长记忆和厚尾特征,在通常的市场条件下,资产收益率出现连续变化,然而在国家宏观调控政策发布日和公司利润发布日等重要日子,波动率表现出快速大幅增加的不连续变化

【参考文献】