| Followed by the openning of the financial market of china,it is facing more and more serious market risk,so it must quickly establish the morden system of market risk management in china.The most important of the market risk management is to measure the risk precisely. for the measuring of the market risk ,VaR introduced by the JP.Morgan in 1994 is the mos popular tool at present. But as the development of the risk management technical, more and more people find that the hypothesis of the traditional VaR model is not suitable for the emerging financial market.on the basis of tracking the newest development of risk measure,this paper analyze the shortage and the applicability of the tradional VaR model, then especially analylize the feature of Chinese security market.as a result ,so as to test that our financial market is suitable for the hypothesis of VaR model or not. As a result, it find that our market is very different from the hypothesis of the traditional VaR model.So the paper introduce the expected shortfall model for measuring the market risk in our financial market ,and introduce two semi-parameter method for calculating VaR ,one is based on the extreme value theory ,the other is based on the quantile regression model.furthermore,through construct a portfolio model ,we compared the improving VaR model withtradionl VaR,so we could find the serious ourcome and shortage of the tradional VaR model directly.In the same time ,whether the expected shortfall method or the VaR,calculating the VaR precisely is critical ,so we choose the newest daily dealing data of four representive security index in the Shenzhen security exchange and the Shanghai security exchange for empirical analysis .in the last,we find that the calculating method based on the extreme value theory is better than others when the interval level is high.and the quantile regression method have no advantage over the others method when we choose the simplicity formality. |
PDF Full Text Request