科研之友
微信
新浪微博
Facebook
分享链接
摘要
In this paper, we propose a novel macroeconomic forecasting model based on the revised multimedia assisted BP neural network model and the ant colony algorithm. Macroeconomic forecasting foundation forecasts the object past and present operating law, therefore, when the operational predict that must describe the analysis and this rule. Because the limitation and forecast technique of choice fault forecast technique can create the uncertainty of the forecasting result, our model mainly focus on the following two aspects. (1) Uncertainty in forecasting method selection errors is even more evident. The probability that the wrong prediction method brings the correct prediction result is very small. (2) Limitations of the forecasting methods. Any kind of forecasting method has its applicable conditions and the environment, it is not omnipotent, nor is it immutable, therefore, more of the state-of-the-art techniques should be researched to enhance the traditional approaches. We use the ant colony algorithm to modify the BP model to make it fit for holding the character that forecasting that a point refers to forecasting a definite value, this value and actual value completely same possibility is very low, this explained that a point forecast successful probability is very low, therefore uses the forecasting result judgement forecast method the fit and unfit quality to be not very comprehensive. Forecast that a sector refers to the future reality leaving in the prediction interval, or prediction interval including the future realistic value which will hold special meaning. The experiment on the stock, gold, exchange and inflation indicate that the proposed model can predict the price well with the satisfactory result.
- 出版日期2017-9
- 单位湖南女子学院
