The Application Of Value At Risk In Risk Control Field

With the development of world economy, the trend of economic globalization and financial integration is becoming increasingly apparent, financial technology continues to develop, the world financial market and financial environment has undergone enormous changes, financial risks continue to increase, then financial risk management has emerged along with increased financial market risks, becoming the principal content of financial enterprises and business administration agency. In order to better measure of risk, after a lot of analysis of national financial community, based on mark-to-market, Value at Risk is becoming the main method to measure risk of financial community.The method of VaR is a very popular international risk management tool in recent years, it has a very wide range of applications in financial risk measurement, prediction and control, and other fields. The most important content of VaR is to deal distribution Fitting and tail of the distribution, there is no doubt this problem also is one of the most popular questions in foreign financial field. Traditional studies suggest that changes in financial market yields variables should obey the Gaussian distribution, however, recent research suggests that Traditional Gaussian distribution model seriously underestimated the risks, because the distribution of the return of economic variables in reality is not strictly obey Gaussian distribution, it Similar to Gaussian distribution, but has thick tail effect, that is the tail of the distribution is more hypertrophic than the tail of Gaussian distribution. About how to deal thick tail effect, there isn’t a uniform standard method, someone used Generalized Pareto distribution to instead the normal distribution, someone by improving the traditional historical simulation to calculate VaR, in short, there is not has a uniform standard method to calculate VaR.In this paper, at first we will introduce the origin and definition of VaR, then discuss the main methods used to calculate VaR in academia and application community, by analysis and comparison this methods, we will give the advantages and disadvantages of each method, and give the improvement program of calculating VaR, make sure our method of calculating VaR can accurately measure the risks. However, in the improvement of VaR, we will introduce other mathematical models like EWMA model and GARCH model which will used in updating the underlying variable’s volatility, then we will give introduce in detail. Finally, we will do the empirical analysis, by VaR system or VaR model to calculate the portfolio’s VaR, and used the historical data to do the review test, compare the actual daily losses of our portfolio whether exceeded the portfolio’s VaR which calculated by our VaR model or not, at this time, if the actual daily losses of our portfolio exceeded the portfolio’s VaR, we call this case is exceptions. For example, if the confidence level of our portfolio’s VaR is99%, then in the review test, we need to test the proportion of exceptions occurred whether reach1%or not, if the proportion of exceptions occurred less than1%, then the VaR which calculated by our VaR model is accurate, otherwise, our VaR model is not accurate, at this time, we need to adjustment the method of calculating VaR to improve the accuracy of our VaR model. Review test is a very important part of our VaR system, we may encounter many problems in this process, then we will discuss this problem in detail in this paper to ensure the accuracy of our VaR system...