| Since the reform and opening up,with the continuous improvement of market mechanisms,China's capital market has a greater development.As the main way of reforming state-owned enterprises,the stock market has become an important place for state-owned enterprises to carry out mechanism reform and financing.The stock market not only played an irreplaceable role in promoting the sustained growth of the national economy,but also had great influence on the world economic integration.Compared with the traditional market,the stock market is full of uncertainty in the market,information disclosure,capital flows will lead to fluctuations in the stock market prices,so that investors are difficult to control.Relatively more intense stock price volatility has increased the investor's investment risk and brought losses to investors.Therefore,the study of stock market price volatility is very important and very meaningful.In this paper,we use mathematical tools and measurement methods to describe and forecast the characteristics of China's securities market,to reveal its inherent rule.Through the analysis of stock price fluctuation rate,it will deepen the understanding of the fluctuat io n of stock price volatility,grasp the trend of future stock market,and find out the possible problems of stock market in our country and put forward the corresponding improveme nt measures.Furthermore,by comparing the theoretical results with the actual results,we find the place where the theory and practice are inconsistent,and then improve the defects of the model we used.In the descriptive statistical analysis,in addition to its main indicators,including the mean,standard deviation,skewness,kurtosis analysis,but also found that the fluctuations are aggregated and cyclical.As can be seen from the results,the yield sequence presents a characteristic of sharp peak and heavy tail.The realized measure of different frequencies also presents different characteristics.The smaller the time interval,the larger the frequency and the smal er the mean of the sequence.Among them,the realized variance at the 5-minute interval is very close to the characteristics of the realized kernel estimate.For the aggregat io n of the yield series,that is,it will follow another large fluctuation after a larger fluctuat io n,and will follow another smal er amplitude fluctuation after a smaller fluctuation.Its cyclica l,mainly reflected in the macroeconomic cycle and consistent.In the empirical analysis of Skewed-T Realized GARCH model,we establish the best model hysteresis order and optimal distribution by comparing the goodness of the data with the model,and also use the information impact curve to analyze the asymmetric effect of volatility.In the analysis of the prediction ability of the model,the advantages of Skewed-T Realized GARCH are verified by comparison with Skewed-Normal Realized GARCH and traditional EGARCH model respectively from both in-model prediction and out-of-model prediction.The advantage of the Skewed-T Realized GARCH model lies in the newly introduced realized measure,which is due to the intensity of the volatility itself.Whereas the traditional EGARCH model is a simple ARMA process at the yield level,to which rapidly changing volatility sequence is weakly fitted.However,the realized measure can react quickly to changes in intraday volatility. |
PDF Full Text Request