主管：中国科学院
主办：中国优选法统筹法与经济数学研究会
　　　中国科学院科技战略咨询研究院
中国管理科学

中国管理科学 ›› 2011, Vol. 19 ›› Issue (4): 144-151.

• 论文 • 上一篇    下一篇

无偏GM(1,1)幂模型其及应用

王正新1, 党耀国2, 练郑伟2   

  1. 1. 浙江师范大学经济与管理学院, 浙江 金华 321004;
    2. 南京航空航天大学经济与管理学院, 江苏 南京 210016
  • 收稿日期:2010-06-07 修回日期:2011-06-06 出版日期:2011-08-30 发布日期:2011-08-30
  • 作者简介:王正新(1981- ),男(汉族),江苏高邮人,浙江师范大学经济与管理学院,讲师,博士,研究方向:预测与决策方法、科技创新管理.
  • 基金资助:

    国家自然科学基金项目(71071077);全国教育科学“十一五”规划青年课题(EIA100402)

Unbiased GM(1, 1) Power Model and Its Application

WANG Zheng-xin1, DANG Yao-guo2, LIAN Zheng-wei2   

  1. 1. School of Economics and Management, Zhejiang Normal University, Jinhua, 321004, China;
    2. College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Received:2010-06-07 Revised:2011-06-06 Online:2011-08-30 Published:2011-08-30

PDF (PC)

1343

被引次数

64

摘要： 基于GM(1,1)幂模型的模拟误差分析,本文提出了无偏GM(1,1)幂模型及其参数优化方法.从理论上证明了无偏GM(1,1)幂模型对传统GM(1,1)幂模型及其本身的时间响应函数所表达的曲线进行模拟和预测具有重合性,其参数优化方法可以准确识别原始数据所蕴含的参数特性,完全消除了GM(1,1)幂模型自身固有的偏差.其建模过程避免了传统方法由差分方程向微分方程的跳跃导致的误差,应用范围覆盖了无偏GM(1,1)模型和离散灰色模型.数值模拟和实例分析表明,无偏GM(1,1)幂模型使得传统模型的模拟与预测精度得到了显著的改善.

关键词: 灰色系统, 无偏GM(1,1)幂模型, 参数优化, 预测

Abstract: Based on the analysis of inherent errors in traditional GM(1,1) power model,this paper puts forward Unbiased GM(1,1) power model.It is theoretically proved that the complete coincidence of the prediction and simulation to the original data of its time response curve generated form has achieved.The unbiased model presented in this paper has completely eliminated the inherent simulant error of the traditional model,but also avoided the jumping errors from the differential equation to differential equation in traditional grey modeling.The application of this model covers unbiased GM(1,1) model and gray discrete model.Numerical simulation and case analysis shows that the simulation and prediction accuracy in traditional modeling has been significantlg improved by unbiased GM(1,1) power model.

Key words: grey system, unbiased GM(1, 1) power model, parameters optimization, forecasting, forecasting

中图分类号: