中国金融业不同板块间风险传导的非对称性研究——基于非对称MVMQ-CAViaR模型的实证分析

发布时间：2018-03-06 23:02

  本文选题：风险传导　切入点：MVMQ-CAViaR模型　出处：《中国管理科学》2017年08期 　论文类型：期刊论文

【摘要】：针对原始MVMQ-CAViaR模型未考虑正负冲击具有非对称性的不足,本文将其扩展为非对称MVMQCAViaR模型和联合非对称MVMQ-CAViaR模型,继而运用该模型分析了我国金融业不同板块间的风险传导效应,并采用严谨的后测检验对比了各个模型的预测效果。结果表明,银行对证券和保险板块均具有显著的风险传染效应,而证券只能单方向地吸收其他板块的风险溢出;正负信息冲击对自身以及其他板块存在不同程度的非对称特征,且指数下跌对VaR的影响效果要强于指数上涨,联合负向冲击会放大原有的风险水平;新构建的两个非对称模型能显著提升原有模型风险预测精度,其中联合非对称MVMQ-CAViaR模型的预测效果更佳。
[Abstract]:In view of the disadvantage that the original MVMQ-CAViaR model does not consider the asymmetry of positive and negative shocks, this paper extends it to asymmetric MVMQCAViaR model and joint asymmetric MVMQ-CAViaR model, and then uses the model to analyze the risk conduction effect between different sectors of the financial industry in China. The results show that banks have significant risk contagion effects on securities and insurance plates, while securities can only absorb risk spillover from other plates in one direction. Positive and negative information shocks have asymmetric characteristics to themselves and other plates, and the effect of index decline on VaR is stronger than that of index rising, and the combined negative impact will amplify the original risk level. The two new asymmetric models can significantly improve the risk prediction accuracy of the original model, and the combined asymmetric MVMQ-CAViaR model is more effective.
【作者单位】： 复旦大学经济学学院;华中科技大学经济学院;中南财经政法大学;
【基金】：中央高校基本科研业务费资助项目(2016AD007) 国家自然科学基金资助项目(71402191)
【分类号】：F224;F832
，

本文编号：1576853