【主讲】Cathy Ning (Ryerson University)
【主题】Volatility Clustering, Leverage Effect, and Copulas
【时间】2015年5月15日 (周五) 15：30-17：00
【地点】上海财经大学经济学院楼701室
【语言】英文
【摘要】This research investigates the volatility clustering and leverage effect of asset returns with copulas. Volatility clustering and leverage effect are both well-known stylized features of financial asset returns, which are mainly captured in a GARCH model and its extensions. In this research, we use vine copula models to capture the possible nonlinearity and asymmetry in these features. Ning, Xu, and Wirjanto (2015) employ an univariate copula approach and find nonlinear and asymmetric volatility clustering in asset returns, i.e., clusters of high volatility are much stronger than clusters of low volatilities. Models accommodates asymmetric volatility clustering can significantly improve the out-of-sample forecasts of Value-at-Risk. Ning, Xu, and Wirjanto (2008) use pair copula models allowing for non-Gaussian marginal distributions and document a nonlinear leverage effect. This research extends our previous work and captures both features at the same time via the use of vine copulas for higher dimension dependence and the across time univariate dependence. The advantage of using such a model is that it does not impose a linear structure on the volatility clustering such as a GARCH model and it also allows for nonlinear and asymmetric feedback effect and leverage effect. The more complete vine copula approach modeling both features simultaneously is expected to further improve the forecasts of VaR, which is important in risk management.
