摘要
分析了GM(1,1)模型的构造原理,指出初始值和背景值是GM(1,1)模型建模的关键,利用最小二乘法原理和GM(1,1)模型传统预测公式改进了初始值,并利用GM(1,1)模型建模条件建立了新背景值计算公式,从而建立了既适应低增长指数序列,又适应高增长指数序列的集成GM(1,1)模型,实例计算结果表明集成GM(1,1)模型的模拟精度较原始GM(1,1)模型和单方面改进的GM(1,1)模型有较大提高且适应范围更广,为提高模型建模精度和适应范围提供了一个新的途径。
The structure principle of GM（1, 1） was ananlyzed, the initial value and the background value was pointed out as GM（1, 1） model key.Based on the principle of the Least squares method and GM（1, 1） model traditional prognostic formula, the paper has built the new initial value, and has established the new background value formula using GM（1, 1） model modelling condition, then the paper has made a improvement to GM（1, 1） model, established both the adapted low growth index series model, and the adapted high growth index series GM（1, 1） model.the example computed result indicated that the integrated GM（1, 1） model proved the simulation precision and the adaptation scope.the paper gives new way to improve the prediction accuracy and the application scope.
出处
《长春理工大学学报（自然科学版）》
2008年第2期141-144,共4页
Journal of Changchun University of Science and Technology(Natural Science Edition)
基金
吉林省教育厅科技计划资助项目
关键词
初始值
背景值GM(1
1)
模型
集成
initial value
background value GM（1, 1）
model
integration
