Financial risk measurement based on Copula theories

Financial risk management draws more and more attention with the global economic integration.The current studies show that the portfolio diversification can reduce risk to some extent.Most traditional risk measurement models are based on normal distribution,which is always inconsistent with the facts,especially when some extreme events occur.Therefore,the asymmetric GARCH-GJR model is introduced.Using the copula function,GARCH models and the extreme theory,this paper establishes the copula-GJR(1,1)-EVT model to measure the risk of the portfolio of 10 Shenzhen stock index under the equal weight,and compares the results of the single-stock index and the portfolio with the results of the portfolio based on the Gaussian Copula and t-Copula function.