The Application And Some Discussion Of Combination Forecast About Chinese Stock Market

Combination forecast is an important field in forecast. In this paper we focus on the weighting method of combination forecast, and put forward a new weighting method. We analysis the Chinese stock market's history data with the combination forecast theory, statistical analysis, econometrics and gray system theory, and exploit some models to forecast our stock price tendency. Empirical analysis shows that the combination forecast can improve the forecast accuracy greatly. At the same time, the new weighting method also can improve the forecast accuracy with more simple calculation and its forecast accuracy is almost the some as optimal combination forecast. In this paper, we focus on the weighting method of combination forecast. Our main work in this paper lies in as follow: first we systemically explain the development of combination forecast, and introduce the common knowledge on forecast, based on this we expatiate the combination forecast---from its origin to development, from its shortcoming to advantage and so on. Secondly we put forward a new weighting method which has practicability; in our empirical analysis, we forecast the stock price by three different methods, namely Regression analysis, Gray forecast and time series analysis, and associate with the three method to a combination model. The result shows that the combination forecast can improve the forecast accuracy greatly. At the same time, the new weighting method also can improve the forecast accuracy with more simple calculation and its forecast accuracy is almost the some as optimal combination forecast. Thirdly we make a new attempt by exploiting the combination forecast to our stock market. At last we make a summary of the paper and point out the farther evolution on combination forecast.