首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于离散指数函数优化GM(1,1)模型的再优化
引用本文:薛焕斌,魏勇.基于离散指数函数优化GM(1,1)模型的再优化[J].数学的实践与认识,2009,39(1).
作者姓名:薛焕斌  魏勇
作者单位:西华师范大学数学与信息学院,南充,637002
基金项目:四川省教育厅重点项目,四川省应用基础研究项目 
摘    要:分析了基于离散指数函数优化的GM(1,1)模型虽然大幅度提高建模的精度,但在构造新背景值过程中仍存在误差的原因,并针对此原因提出了进一步优化此背景值的方法,从而再次提高了建模的精度.经过严格理论验证该模型具有白化指数重合性,所以既适合用于低增长指数序列建模,也适合用于高增长指数序列建模.同时通过大量的数据模拟,并与原GM(1,1)模型及其基于离散指数函数优化的模型对比,发现本文优化的GM(1,1)新模型有非常高的模拟精度和预测精度.

关 键 词:GM(1  1)模型  背景值  优化

Optimize an Optimal GM(1,1)Based on the Discrete Function with Exponential Law Once Again
XUE Huan-bin,WEI Yong.Optimize an Optimal GM(1,1)Based on the Discrete Function with Exponential Law Once Again[J].Mathematics in Practice and Theory,2009,39(1).
Authors:XUE Huan-bin  WEI Yong
Abstract:This paper analyzes the reason why there exist a error during structuring a new background value in an optimal GM(1,1) based on the discrete function with exponential law,although it has improved the modeling precision greatly,and put forward a further optimization method for this reason.Then,obtain a new GM(1,1) model and improve the model precision further.The new model has been proven strictly to have the property of white exponential law coincident,so it not only to be suitable for the low growth sequence,but also suitable for the high growth sequence.Through simulation to a large number of data,and compared with the original GM(1,1) model and the optimal GM(1,1) based on the discrete function with exponential law,then we discovered that the new optimized model in this paper has very high simulation and forecasting precision.
Keywords:GM(1  1) model  background value  optimization
本文献已被 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号