Chinese Stock Market Risk Based On The Var Historical Simulation Method Study

The financial markets have been developing rapidly in recent years, accompanied by the unprecedented strong fluctuation. The growing risk of financial markets requires the high attention on its monitoring and measurement. In past two years, the stock market in China fluctuated significantly but attracted a lot of investors. However, without paying close attention to the potential risks many investors had suffered heavy losses. Many people know little about risk measurement of the stock market. In return, their blind investment might exacerbate the stock market fluctuations and risk. Therefore, it is necessary to introduce and promote an easier and convenient method of measuring risk in the stock market, which is helpful for people to measure risk and invest rationally. This will also benefit the healthy and stable development of China's stock market.VaR is a new risk measurement technology, which has been widely used by official government and financial institutes due to its concise, comprehensive, and practical features. VaR is constantly being optimized and becoming the prevailing method for managing market risks.In the third section, the VaR method is been introduced systematically, including the principle of calculation methods, advantages and disadvantages, and etc. Later, the historical simulation of VaR is used to analyze and validate the risks in different periods of Shanghai and Shenzhen stock markets. The result showed that in the stable market, historical simulation can estimate the risk very well. However, it can not truly reflect the status of market risk when there are large fluctuations in the market. Based on this finding, this paper compared two periods of data distribution, and optimized the data obtained from historical simulation. This new proposed method is proved to be very good for risk analysis when the market is not stable. In addition, the historical data were also processed with exponential smoothing, and then applied to historical simulation method to calculate the stock market risk. The results showed that the smoothed data was better than the original data to be used as input in historical simulation in some cases, which lead to more precise risk estimation. Finally, this work also provided some policy recommendations based on previous conclusion.