Study On The Short-term Combined Forecast Modeling Approach In Chinese Stock Market By Grey Theory

With the development of social economy and the improvement of people’s income level, the stock has become an important tool for people to invest and finance. But the security market is still in the initial stage in China, its volatility and risk are far higher than the foreign mature security market. So predicting the share price accurately for investment decisions has very important instruction meanings. This paper uses the grey system to the stock market based on the researches of grey forecasting models, at the same time, in order to improve the precision of grey model, it will combine grey system with the SVM theory for the short-term forecast of the stock market. The main content is as follows:In chapter1, it mainly introduces the stock prediction method and summaries the security investment analysis method, mathematical statistics method, modern technical analysis method, grey system theory, and so on.In chapter2, this part firstly discusses the Chinese stock market which does not comply with the random walk model, and the share prices exist the fluctuation, and then explain the Chinese stock market does not meet the weak-form efficiency, and to a certain extent prove the stock market can be predicted in China.In chapter3, this part firstly introduces the grey system theory, then according to the problem that the traditional GM(1,1) simulation sequence does not reflect the dynamic change on the smooth ratio and stepwise ratio of the original sequence, this paper is based on the definition of the smooth ratio and stepwise ratio, establishing the GM(1,1) combination forecast model based on the sequence of smooth ratio and stepwise ratio, the empirical analysis shows the validity of the new model. According to the stock market which exists suspension and short holidays, trying to use the unequal interval grey GM(1,1) model to predict, the model optimizes the background value and the grey derivative which is substituted by weighted average of forward difference quotient and backward difference quotient, and the accumulating method is applied into the parameters’estimation, the empirical analysis shows the optimized model makes simulation and forecast which has a high precision. At last, analyzing the principle of GM(1,N) model, and indicating that the classical GM(1,N) model has three shortages, then putting forward the improvement method and establishing the optimization method of GM(1,N) model. The empirical analysis shows that the validity of the new model, it can be used to predict the short-term share price.In chapter4, we combine grey system theory with SVM method, and put forward the GM(1,1) model based on SVM and the GM(1,N) model based on SVM, then according to the definition of the smooth ratio and stepwise ratio, establishing the nonlinear grey model based on SVM by the sequence of stepwise ratio and the nonlinear grey model based on SVM by the sequence of smooth ratio and stepwise ratio. The empirical analysis shows that combination of the grey system theory and SVM method can be well applied to forecasting stock price, which has a high precision.In chapter5, this part introduces the main research contents, research results and innovation points, also discussing the future research work.