VaR Method And Its Application In Risk Management Of China's Stock Market

Since the 70's of last century, the finance risk management technology has developed swiftly and violently under the financial turbulent pressure. Because of its simple and clear advantages, VaR is seen as the popular method. After briefly introduces the background and the research present situation, the paper firstly explained three kinds of financial risk measurement method: Sensitivity analysis method, Volatility analysis method and VaR method, and compared the three methods; Then detailed instruction VaR method, After compared the various methods choose parameter calculation method to calculate VaR. After that the text introduced the kind model of GARCH in detail. In the last place, the paper comes to the following conclusions by empirical research.This thesis has two traits as following: firstly, the thesis completely introduce the kind model of GARCH, it covers in detail the essential 11 kinds, and Application 8 kinds: GARCH(1,1) model, EGARCH(1,1) model, PARCH(1,1) model, Component ARCH(1,1) model, and the kind model of GARCH-M:GARCH(1,1)-M model, EGARCH(1,1)-M model, PARCH(1,1)-M model, Component ARCH(1,1)-M model. Secondly, the paper using three kinds of distribution: The normal distribution, t-distribution and generalized error distribution.Its conclusion is that GED distribution is better than The normal distribution and t-distribution, and Among all kinds of GARCH model, PARCH(1,1) model is the best.