基于结构突变的波动率分解模型

Based on Structural Breaks Volatility Decomposition Model

在Heston-Nandi模型的基础上提出了一种波动率分解模型,分解模型同时考虑了金融波动的长记忆性和杠杆效应.从资产收益率的无条件方差发生结构突变出发,认为收益率的无条件方差随时间变化,将波动率分解为长期影响和短期冲击两部分,其中长期影响用来刻画波动率的持续性,短期冲击刻画金融波动的短期扰动.上证综指数据实证表明上海证券综合指数收益率序列的波动性同时具有长记忆性和杠杆效应,利用模型能很好的刻画这两种性质.

This paper presents a volatility decomposition model based on Heston-Nandi model, decomposition model taking into account the long memory in financial volatility and leverage. In this paper, we assume the un-conditional variance of returns on assets has structure breaks. That means the unconditional variance of returns over time, so, volatility can be divided into long-term impact of volatility and short-term impact, Long-term impact is used to describe the long-term effects of continuous volatility, short-term impact Describe the short-term financial volatility disturbance. Empirical data show that the Shanghai Composite Index also has long memory of volatility and leverage effect, use of our model can well describethis two properties.