摘要
本文提出一个利用混频数据估计资产波动率的框架,该框架使用日内高频数据构造蕴含潜在发生概率的跳跃和扩散波动指标,以外生的滞后项进入回馈函数,既能充分利用样本信息,又能避免无限滞后期的回馈影响。在对沪深300指数的实证分析中,考虑一个跳跃对扩散波动具有非对称性溢出效应的双向波动率回馈模型。相对于基准模型,这一模型对数据的描述更优。分析结果显示,两类波动间存在正向回馈效应:跳跃向扩散的溢出导致自回归条件异方差(ARCH)系数存在两个区制且区制内的变异性明显;扩散向跳跃的溢出致使跳跃强度的自相关性在极端市场环境中出现强化。波动率回馈机制使得信息释放后价格反复调整变化,导致波动率高企;熔断事件折射出A股信息流质量差、融解效率低等问题。由此可以得出结论:相关监管和交易制度亟待完善。
This paper provides a framework of estimating asset volatility with mixed-frequency data,using the intraday high-frequency data to construct diffusion and jump indicators containing potential occurrence probabilities,which enter the feedback functions as exogenous lagged terms.This method can not only make full use of sample information,but also avoid the feedback from the infinite lagged period.In the empirical analysis applied to the CSI300 index in China,the feedback function is set concretely to consider a model with bidirectional volatility feedback where exists the asymmetric spillover from jump to diffusion.Compared with the benchmark models,the new model has a better description of the data and shows positive feedback effects between both types of volatilities.The jump-to-diffusion spillover results in the existence of two regimes in the ARCH coefficient and the obvious variability within each regime,while the diffusion-to-jump spillover causes reinforcing autocorrelation in jump intensity in extreme market environments.The volatility feedback mechanism makes repeated price adjustment after information release,leading to high volatility,and the circuit breaker events reflect the problems of poor information quality and low efficiency of absorption.Therefore,relevant regulation and trading systems need to be improved.
作者
任光宇
吕小锋
REN Guangyu;LYU Xiaofeng(Capital University of Economics and Business,Beijing 100070;Southwestern University of Finance and Economics,Chengdu 611130)
出处
《经济与管理研究》
CSSCI
北大核心
2021年第2期66-81,共16页
Research on Economics and Management
基金
国家自然科学基金青年科学基金项目“跳跃风险与股指期货套期保值:基于低频和高频市场信息的研究”(71401112)
教育部人文社会科学研究一般项目“基于辅助信息的不平等指数研究:方法与应用”(20YJCZH117)
首都经济贸易大学北京市属高校基本科研业务费专项资金资助项目。
关键词
资产收益率
跳扩散
波动率分解
波动率回馈
高频金融数据
asset returns
jump-diffusion
volatility decomposition
volatility feedback
high-frequency financial data
