摘要
研究目标:考虑资产收益率条件高阶矩的动态特征以及突发事件对于金融市场的影响,构建基于结构突变的动态高阶矩Realized EGARCH模型,对创业板市场波动率进行预测和风险度量。研究方法:在Wu等(2020)提出的Realized EGARCH-SK模型的基础上,将残差拓展为偏t分布,同时应用修正的ICSS算法识别结构突变点,将其作为虚拟变量引入动态高阶矩Realized EGARCH模型中,并应用稳健损失函数以及(非)条件覆盖检验来综合比较各模型的波动率预测精度和VaR预测准确性。研究发现:考虑条件偏度和峰度动态效应以及结构突变因素的Realized EGARCH模型有助于提高模型的拟合能力、波动率的预测精度和风险度量的准确性;Gram-Charlier扩展分布在模型的拟合能力和VaR预测的表现上明显不如偏t分布。研究创新:将残差拓展为偏t分布同时加入结构突变因素有效改进波动率模型的拟合、预测和风险度量效果。研究价值:丰富了动态高阶矩高频波动率模型,结构突变的监测和纳入可为金融投资者和风险管理者的投资决策提供实质性参考。
Research Objectives:Considering the dynamic characteristics of the conditional higher moments of asset return and the impact of emergencies on the financial market,constructing the dynamic higher moments Realized EGARCH model based on structural breaks to predict and measure the volatility of the GEM market.Research Methods:Based on the Realized EGARCH-SK model proposed by Wu et al.(2020),the residual is extended to the skewed-t distribution,and the modified ICSS algorithm is used to identify structural breaks,which are regarded as dummy variables and introduced into the dynamic higher moments Realized EGARCH model,and the robust loss function,unconditional coverage test and conditional coverage test are used to comprehensively compare the volatility prediction accuracy and VaR prediction accuracy of each model.Research Findings:The Realized EGARCH model,which considers the dynamic effects of conditional skewness and conditional kurtosis and structural breaks factors,can help to improve the fitting ability of models,the prediction accuracy of volatility,and the accuracy of risk measurement.The Gram-Charlier expansion distribution is obviously inferior to the skewed-t distribution in the fitting ability of the models and the performance of VaR prediction.Research Innovations:Extending the residual to the skewed-t distribution and adding structural breaks factors to effectively improve the fitting,prediction and risk measurement effects of the volatility models.Research Value:It enriches the high frequency volatility model of dynamic higher moments,and the monitoring and incorporating of structural breaks can provide a substantial reference for the investment decisions of financial investors and risk managers.
作者
蔡光辉
廖亚琴
Cai Guanghui;Liao Yaqin(School of Statistics and Mathematics,Zhejiang Gongshang University)
出处
《数量经济技术经济研究》
CSSCI
CSCD
北大核心
2021年第1期157-173,共17页
Journal of Quantitative & Technological Economics
基金
国家社科基金2019年“高频金融数据统计测度模型的拓展研究”(19BTJ013)的资助
