Describing And Forecasting Of Security Market Return Based On Its Distribution Characteristics And Macroeconomic Factors

The security market return is an important conception in the area of the financial economics, and the accuracy of the description of the return volatility generally affects the correctness of the security portfolio, the validity of the risk management and reasonability of the option pricing. For one side, as a complex system, security market has some instinct running pattern, thus the return follows certain distribution characteristic, consequently, it is essential to consider the instinct running pattern to describe and predict the return of the security market. For the other side, the running of the security market is undoubtedly shocked by the external information and other factors. Among these factors, the macroeconomics which contains series of variables is the most direct and influencing factor, therefore, in the process of the describing and predicting the security return, it is necessary to consider the instinct running rules and the mechanism of the external macroeconomic factors.The thesis is to explore the new ways to describe and forecast the security return combining the internal factors and the external factors of the security market development. By integrating effects of the distribution characteristics of the security market return and the macroeconomic variables, the thesis explores a more efficient forecasting combination model to predict the trend of the security market and the instinct mechanism of the security market.The thesis firstly empirically examines the effectiveness of the Chinese security market, and finds the security market has achieve the weak form efficiency, but there is still a long way to reach the semi-strong form efficiency. In Chinese security market, inside investors often get abnormal return by using the inside information, while medium and small investors doesn't acquire the deserved return in the market.To authentically reflect the running pattern of the Chinese security market, the thesis adopts different methods to analyze the distribution characteristics of the security market in Shanghai Stock Exchange and Shenzhen Exchange. The results show three outstanding characters: these two stock markets are closely correlated; the overall day return doesn't follow the normal distribution, but it is characterized with leptokurtic and fat tail; the security market return has stable characteristics and multi-fractal characteristics which shown as the long memory, fat tail, scale properties, and variability. The long memory of the security market return means that the predicting ability can be enhanced by adopting more history information. The thesis deeply studies the long memory characteristic and its causes of the security market in China. Meanwhile, choosing composite index of Shanghai stock exchange and component index of Shenzhen stock exchange to represent the overall change trend of the markets, the long memory characteristic of the security market returns is described by using different method, and the appropriate models are found by the empirical study.Security market is the outcome of the development of the economics, thus the status of the security market is closely connected with the economic development of the country. This thesis theoretically and empirically studies the effects of macroeconomic on the security market return, and the results indicate that there exists a long run equilibrium relationship between security market return and various macroeconomic variables. The development of security market is consistent with the development of macro economy in principle, and the short-run fluctuation of security market return is affected by added industry value, money supply, inflation rate, interest rate, the short run change of the saving and so on.The government policy is the important factor that affects the security price, security return and the volatility. The thesis analyzes the implication of China's Policy Market and the cause of its formation, explains the China's Policy Market by taking the contingent governance model, and empirically detects the market effect of the important polices.Lastly, considering the distribution characteristics and the long memory characteristics, the monthly return of the Chinese security market is predicted by using GARCH models. Similarly, considering the effects of the macroeconomics on the return of the security market, the monthly return of the Chinese security market return is predicted by taking VAR models. Meanwhile, integrating the distribution of the security market return and the effects of the macroeconomics, VAR-GARCH models is build to prescribe and predict the security market return. By comparing the predicting effects, the VAR-GARCH models are superior to the other models.