基于分段线性Lagrange插值Grey-Markov模型的CPI预测  

CPI Prediction Based on Piecewise Linear Lagrange Interpolation Grey-Markov Model

作　　者：郑雪平[1] 水庆象 Zheng Xueping;Shui Qingxiang(School of Economics and Management,Southwest University of Science and Technology,Mianyang Sichuan 621010,China;School of Human Settlement and Civil Engineering,Xi’an Jiaotong University,Xi’an 710054,China)

机构地区：[1]西南科技大学经济管理学院,四川绵阳621010 [2]西安交通大学人居环境与建筑工程学院,西安710054

出　　处：《统计与决策》2020年第14期29-32,共4页Statistics & Decision

基　　金：四川省科技厅软科学资助项目(2018ZR0244,2017ZR0206);四川省教育厅资助项目(17ZA0398);四川省军民融合研究院资助项目(18sxb003,2017SCII0106,18sxb007);西南科技大学研究生创新基金资助项目(17ycx065)。

摘　　要：针对传统GM(1,1)模型在背景值构造方面的不足导致预测波动型数据时,预测精度较低的问题,文章提出了Markov修正的分段线性Lagrange插值GM(1,1)预测模型。该模型通过采用分段线性插值法和Lagrange插值法相结合的组合插值思想对背景值进行优化,进一步采用Markov模型预测结果进行修正。通过对具有波动性质的CPI数据进行预测,表明该模型可以削弱数据波动对预测精度的影响,提高模型的适用性,Markov进行修正又可再次提高模型的精度。The deficiency of the traditional GM(1,1)model in background value construction leads to the prediction of fluctuating data,hence the low prediction accuracy.To solve this problem,the paper proposes a GM(1,1)prediction model of piecewise linear Lagrange interpolation modified by Markov.The proposed model optimizes the background value by using the combined interpolation thought of piecewise linear interpolation and Lagrange interpolation,and then corrects the prediction results by adopting Markov model.The prediction of CPI data with fluctuation property shows that the proposed model can reduce the influence of data fluctuation on the prediction accuracy,with its applicability improved,and that Markov's correction can increase the accuracy of the model again.

关 键 词：GM(1 1)模型 背景值优化 MARKOV模型 波动型 

分 类 号：F224.9[经济管理—国民经济]