CAJ | 학술논문

本文以香港恒生指数、德国法兰克福DAX指数和美国S＆P500指数为对象，将三个股指收益组成资产组合。分别以次贷危机和欧债危机爆发为界限，将样本划分为三个时间区间。基于时变SJC-Copula-EVT模型，分别构建VaR和Es风险模型，并通过后验分析方法对比研究风险模型在各个时段的测度精度。研究表明，危机爆发后，VaR模型对资产组合多头头寸的风险测度精度有所提高；而时变SJC-Copula-EVT-ES模型则对资产组合极端风险测度表现出良好的预测效果。
본문이향항항생지수、덕국법란극복DAX지수화미국S＆P500지수위대상，장삼개고지수익조성자산조합。분별이차대위궤화구채위궤폭발위계한，장양본화분위삼개시간구간。기우시변SJC-Copula-EVT모형，분별구건VaR화Es풍험모형，병통과후험분석방법대비연구풍험모형재각개시단적측도정도。연구표명，위궤폭발후，VaR모형대자산조합다두두촌적풍험측도정도유소제고；이시변SJC-Copula-EVT-ES모형칙대자산조합겁단풍험측도표현출량호적예측효과。
Hang Seng Index in Hong Kong, Frankfurt＇ s DAX Index of Germany and the U.S. S ＆ P 500 Index are studied in this paper. A portfolio is formed by the three stock index returns, and the sample is divided into three time intervals with the U.S. Sub-prime crisis and the European Debt Crisis. The VaR and ES models are constructed based on time-varying SJC-Copula-EVT for the portfolio of each period of time, and the estimation accuracies of the risk models are compared by back-testing. The main results show that the estimation accuracy of VaR for the long position has improvedafter the crisis, and the time-varying SJC-Copula-EVT-ES model produces more accurate predictions for extreme market risk.