| With the development of economy,the opening degree of financial market is increasing gradually.Under this circumstance,all kinds of capital flow rapidly all over the world,and the close trading behavior among financial markets leads to the more obvious linkage effect.So how to avoid risks quickly and effectively has become a more and more concerned problem for investors.Therefore,in order to disperse risks and reduce losses,it is particularly important to solve the portfolio of financial assets and measure risks.This paper is divided into two parts.In the first part,we use Copula-Quantile Regression Model to solve the optimal portfolio scheme between stock data.Firstly,three single Copula functions(Gumbel Copula function,Clayton Copula function,Frank Copula function)and mixed Copula function are used to fit the selected stock data(Hengrui Medicine,AIA).Secondly,select the Copula function with the best fitting effect among the four Copula functions,then Monte Carlo simulation method and the Quantile regression model are used to find the total profit rate of the target stock in the next period for each weight combination.Finally,we test the total rate of profit for each weight combination by Omega ratio to acquire the optimal portfolio.In the second part,we use QR-GARCH-N model to measure the risk of the optimal portfolio scheme.To begin with,obtaining risk measure values of each stock under different confidence levels is necessary.Then we calculate the risk measure value of the portfolio plan according to the characteristics between VaR value and CVaR value.The results show that the mixed Copula function has the best fitting effect when the holding period is 1 day and 15 days,while the Gumbel Copula function has the best fitting effect when the holding period is 30 days.in addition,the optimal portfolio scheme under each holding period and the VaR value and CVaR value of each portfolio plan under different confidence levels are obtained,which investors can used as reference. |
PDF Full Text Request