提高GM(1，1)模型拟合预测精度的研究

Research on Improving the Fitting and Prediction Precision of the GM (1,1) Model

在简单介绍GM(1,1)模型预测过程的基础上,指出了模型在求解微分方程时已知条件选取和背景值构造两方面存在不足,并对此提出了更换已知条件及通过求解最小值获取背景值构造形式的改进措施.通过把这两种措施进行有机结合,形成了一种新的预测程序和方法.并且通过实例分析证明了这种新的程序和方法的确能够提高GM(1,1)模型的拟合预测精度.

For the unreasonable adoptions of the initial known condition and background value when solving the differential equation in the GM （1,1） model, the paper brings forward both the methods of changing the initial known conditions and getting the optimal structure format of the background value through the minimum value theory. And based on the organic integration of the two improving perspectives, a new kind of prediction process and method comes into being. Furthermore, the example analysis at the end of the paper proves that the very new method is certain to improve the fitting and prediction precision of the GM（1,1） model.