金融时间序列系统风险度量:动态二元Dvine模型  

Measuring systemic risk for financial time series:A dynamic bivariate Dvine model

作　　者：陈昱[1] 曹心怡 金姝玥 徐涛 Yu Chen;Xinyi Cao;Shuyue Jin;Tao Xu(Department of Statistics and Finance,School of Management,University of Science and Technology of China,Hefei 230026,China)

机构地区：[1]中国科学技术大学管理学院统计与金融系,安徽合肥230026

出　　处：《中国科学技术大学学报》2023年第11期1-11,I0001,I0007,共13页JUSTC

基　　金：supported by the National Social Science Fund of China(22BTJ027)。

摘　　要：在金融市场中,对金融资产的尾部风险的精准度量一直是研究者关注的焦点。本文提出了一个新的二元时间序列模型,用来计算和预测金融资产的在险价值(VaR)和条件在险价值(CoVaR)。该模型可以同时捕捉二元时间序列中存在的序列相关性和横截面相关性,从而提高估计和预测的精度。本文在模型推导中给出了该二元时间序列模型的参数估计值,并且基于plug-in方法给出了VaR和CoVaR的估计值。还建立了Dvine模型估计量的渐近性质。对金融股价的实证分析表明我们的模型在风险度量和预测方面表现良好。Accurate measurements of the tail risk of financial assets are major interest in financial markets.The main objective of our paper is to measure and forecast the value-at-risk(VaR)and the conditional value-at-risk(CoVaR)of financial assets using a new bivariate time series model.The proposed model can simultaneously capture serial dependence and cross-sectional dependence that exist in bivariate time series to improve the accuracy of estimation and prediction.In the process of model inference,we provide the parameter estimators of our bivariate time series model and give the estimators of VaR and CoVaR via the plug-in principle.We also establish the asymptotic properties of the Dvine model estimators.Real applications for financial stock price show that our model performs well in risk measurement and prediction.

关 键 词：序列相关性 横截面相关性 时变COPULA 金融风险管理 条件分位数 

分 类 号：F830.99[经济管理—金融学]