摘要
首先用含待定系数λ的代数式构造灰色模型GM(1,1)的背景值,然后由最小二乘法优化初始条件,最后在后验差检验的标准下,提出一种邻域收缩法求待定系数λ的值,获得了优化的新模型.实例证明新模型预测精度更高,预测应用范围更广.
In this paper,the background value of the grey model GM(1,1) is constructed by the algebraic formula with undetermined coefficients λ and then the initial condition is optimized by the least square method.Finally,under the standard of the posteriori error test,a neighborhood contraction method is proposed to calculate the value of undetermined coefficients λ,and a new optimization model is obtained.The example shows that the new model has higher prediction accuracy and wider application range.
作者
肖利芳
王浩
段梅
XIAO Li-fang;WANG Hao;DUAN Mei(School of Computer Science and Engineering,Wuhan Institute Technology,Wuhan 430205,China;Key Laboratory of Ntelligemnt Robot in Hubei Province,Wuhan 430205,China)
出处
《数学的实践与认识》
2021年第12期1-10,共10页
Mathematics in Practice and Theory
基金
教育部2019年第二批产学合作协同育人项目(教高司函[2020]6号,201902286002)
2020年国家级大学生创新创业训练计划项目(教高司函[2020]4号,202010490014)
2020年湖北省大学生创新创业训练计划项目(鄂教高办函[2020]9号,S202010490056)。
关键词
GM(1
1)模型
背景值
初始值
后验差
GM(1
1)model
background value
initial value
posteriori error
