StatisticalResearch2023年5月May2023Vol.40,No.5第40卷第5期统计研究基于RT-GASCopula模型的经济金融行业非对称相依性及风险溢出研究徐君郭宝才内容提要:考虑到实体经济行业与金融行业间的相依关系存在杠杆效应,本文采用GJR门限结构将杠杆效应纳入包含高频信息的广义已实现自回归得分Copula(GRASCopula)模型中,构建广义已实现门限自回归得分Copula(RT-GASCopula)模型。利用RT-GASCopula模型揭示实体经济行业与金融行业间时变相依关系对行业收益的非对称响应,并进一步分析经济金融行业间风险溢出的非对称性和时变特征。研究发现:各行业间的相依关系存在显著的杠杆效应,某一行业收益的上涨和下跌对该行业与其他行业间时变相依关系的影响是非对称的;行业间的相依关系会受到国家调控政策和各类市场风险事件的影响;行业间的系统性风险溢出存在非对称性,且会受到严重风险事件的影响。此外,有效性检验结果表明,当行业间相依关系存在显著杠杆效应时,RT-GASCopula模型的拟合和预测能力优于现有时变Copula模型。关键词:相依关系;杠杆效应;GJR门限结构;RT-GASCopula模型;风险溢出D0I:10.19343/j.cnki.11-1302/c.2023.05.005中图分类号:F830文献标识码:A文章编号:1002-4565(2023)05-0064-14ResearchontheAsymmetricInterdependenceandRiskSpilloverofEconomicandFinancialIndustriesBasedonRT-GASCopulaModelXuJun&GuoBaocaiAbstract:Consideringthatthereisaleverageeffectintheinterdependencebetweentherealeconomyandfinancialindustries,theGJRthresholdstructureisusedtoincorporatetheleverageeffectintotheGeneralizedRealizedAutoregressiveScoreCopula(GRASCopula)modelcontaininghigh-frequencyinformationtoconstructtheRealizedThresholdGASCopula(RT-GASCopula)model.Basedonthis,weusetheRT-GASCopulamodeltorevealtheasymmetricresponseofthetime-varyinginterdependencebetweentherealeconomyandthefinancialindustriesonindustryreturns,andfurtheranalyzetheasymmetricandtime-varyingcharacteristicsofriskspilloversbetweentherealeconomyandfinancialindustries.Theresultsshowthat:first,theinterdependenciesbetweenvariousindustrieshaveasignificantleverageeffect,andtheriseandfallofoneindustry'sreturnhaveasymmetriceffectonthetime-varyinginterdependenciesbetweenthatindustryandotherindustries;s...