可变参数动态灰色预测模型的建立与实证研究

在固定参数动态灰色预测模型基础之上进行改进,提出了可变参数动态灰色预测模型.固定参数动态灰色预测模型的参数值为0.5,可变参数动态灰色预测模型的参数值由预测对象的影响因素来决定.根据对我国集成电路产业的应用研究,说明了可变参数动态预测模型具有更好的预测精度.同时,针对可变参数动态灰色预测模型的计算量很大的特点,应用Java编制了应用软件,可以方便快捷地进行运算,有利于该模型的应用推广.

Construction and empirical research of the variable parameter value rolling grey forecasting model

A variable parameter value rolling grey forecasting model(RGM)is constructed,which makes an enhancement to the fixed parameter value RGM.In the fixed parameter value RGM there is a constant parameter value,which is customarily set to 0.5.It is apparent that if the parameter value is decided in consideration of the factors which influence the forecasting objective,better result will be achieved.According to the results of predicting annual Integrated Circuit Industry sales in China,the variable parameter value RGM generates more accurate forecasts.Due to that a lot of calculation work is needed when using this model,an application software is developed taking advantage of the Java for dealing with the processes.The program is of benefit to the application and dissemination of the model proposed.