灰色GM（1,1）模型的改进及其在CPI预测中的应用

针对传统GM(1,1)模型的改进方法复杂、使用范围有限、预测精度不高等问题，本文对传统GM(1,1)模型的背景值进行理论分析并改进，用二次插值的方法重新构造背景值。在此基础上对原始数据通过滑动平均法进行初值预处理，给出改进的模型，最后运用其进行短期预测，仿真结果证明了此改进模型的有效性和可行性，为提高预测精度提供新的途径。

Improvement of GM(1,1) model and its application in CPI

In view of the defects including the difficulty of improvement, limit of application and low accuracy of the traditional GM(1,1) model, we have the theoretical analysis of background values and reconstruct the background values through quadratic interpolation. The initial values are calculated by moving average meth-od on the basis of the reconstructed background values. Finally, the newly improved GM(1,1) model is estab-lished and applied to short-term prediction. The results verify the effectiveness and feasibility of the improved model and give a new way to improve the prediction accuracy.