Methods Of Risk Measurement Estimation And Empirical Study On Financial Time Series Based On Wavelet Analysis

Recent years, the progress of economic globalization, financial integration keeps accelerating, modern financial theory and products innovate rapidly, which makes the global economy gain an unprecedented growth. However, behind this prosperity, global economy is facing increasing financial risks. Especially after several serious global financial crisis, more and more people begin to realize that effective risk measurement method will affect the development of global financial market. Hence, carrying out the research of risk measurement and control of financial time series has extremely significant practical meanings for financial supervising and system security maintaining.Firstly, this paper introduces the method for calculating the VaR, which is an approach to measure financial risk, and analyses the existing problem. Then it mainly focuses on expounding financial time series model and the theory of wavelet analysis. Further, Shanghai composite index closing price from January 2rd, 2014 to December 31 th, 2015 is chosen to perform the model, which is in order to present the methods of risk measurement estimation and empirical study. Specifically speaking, GARCH model is applied to fit the time series of Shanghai composite index closing price under the condition of standard normal distribution and student t distribution respectively, which is used to improve the calculating results of variance-covariance method, which is a general calculating method of VaR. The results show that compared to GARCH(1,1), the failure rate of GARCH-t(1,1) is much more smaller. Further, wavelet analysis is emphatically apply to eliminate the noise of time series, which is aimed to optimize the fitting result of GARCH-t(1,1) and improve the calculating results of VaR. It is shown that the accuracy of prediction is greatly improved. This paper also uses the theory of wavelet analysis to detect the singularity of time series, and the result shows that Shanghai composite index closing price mutated and fluctuated greatly in the middle of 2015, which accords with the fact that the stock market crashed at that time. Therefore, it can be concluded that wavelet analysis has significant application value for the research of risk measurement estimation on financial time series.