多变量非等间距GM(1,m)模型及其应用

对于多变量非等间距数据序列,建立了一类GM(1,m)预测模型。基于灰色模型的指数特性和积分定义,提出了构造多变量非等间距序列的GM(1,m)模型背景值的方法。该方法可以提高GM(1,m)模型的拟合精度和预测精度,拓广了灰色模型的应用范围。应用该方法,建立了中国农村青少年生长水平的灰色预测模型,结果理想可靠,有较好的实际意义。

Multivariable non-equidistance GM (1, m) model and its application

For multivariable non-equidistance sequence,a grey forecast model GM(1,m) is established.Based upon the exponential trait of the grey model and the definition of integral,a construction method of the GM(1,m) model's background value of multivariable non-equidistance sequence is put forward.Moreover,the GM(1,m) model's fitting precision and prediction precision are advanced and the scope of application is enlarged.As the application,a forecast model of growth level of China countrysied teenagers is established.By the realistic example,the result is proved to be precise and reliable.