Calculation Method Research To Value-at-Risk(VaR) Of Stock Market In Extreme Cases

Recently, with the daily volatility of financial market and some financial catastrophic events happened one by one, which put a challenge for risk management, we have been long for some more appropriate models to deal with such events. Because the conventional method to measure is based on normal distribution assumption which has been proved to underestimate risk, in order to measure the risk more accurate, more and more researchers put forward using extreme-value-theory to measure market risk which has been used in engineering field widely. Because extreme-value-distribution need not to put any hypothesis on the whole distribution of return but only by data themselves, fitting the tail of distribution, which is fitted to measure risk. Currently, the Generalized Pare to distribution based on extreme-value theory was widely used to study market risk. In the paper, we prepare to use Generalized Pareto distribution based on extreme-value-theory to study Chinese stock market.It mainly on the basis of their predecessor's Research of the extreme value theory, it discussed how to use extreme value theory to estimate value-at-risk (VaR) of China's stock market more precise. Which is part of the core is that: At present, the application of the EVT theories are generally selected POT mod, one of the key issues in POT model in is the threshold selection, It affect the accuracy of value-at-risk (VaR). Recent research has usually used three methods, one is Expectations of Overruns, and second, Hill plans, and the other is Compared with normal. These three threshold selection method in practical applications have many shortcoming. The paper investigates a new method of threshold selection. According to Glivenko Theory, improves historical simulation to make the calculation result more accurate and use it to determine the threshold of POT model to calculate new VaR of finance market. And taking into account the financial data is not-relation or weak correlation, that is: larger fluctuations will be gathered relatively in a time, and small fluctuations will be gathered relatively in another time. So introduce ARMA-AGARCH model to remove not-relation or weak correlation of data .Then, conducts the empirical analyze using Shanghai index and get a good result.