为解决传统GM(1,1)模型存在的问题,在运用积分中值定理证明含有自适应因子λ∈(0,1)的背景值构造方法可行性的基础上,将该方法引入传统GM(1,1)模型的定义型,推导出了GM(1,1)定义型预测公式,构造了具有自适应能力的GM(1,1,λ)模型。通过理论证明和数据模拟实验两方面的研究结果表明自适应GM(1,1,λ)模型能够克服现存问题,并将模型的适用范围扩大为发展系数a∈(-1/λ,1/1-λ),大于传统GM(1,1)模型的适用范围α∈(-2,2),且可用于高增长序列建模,比传统GM(1,1)模型具有更高的拟合和预测精度。
To deal with the problems existing in conventional GM (1,1) model, the structure method of background value which contained a self-adaptive factor λ∈ (0,1) is introduced into the conventional GM (1,1) model ; s definition based on its feasibility is proved by the mean value theorem for integrals, the prediction formula of GM (1,1) model's definition is deduced, and then, a new self-adaptive GM (1,1, λ ) model is put forward. Finally, both theoretical analysis and results of the data simulation indicate that the self-adaptive GM (1,1,λ ) model can overcome the problems, its applicable region is not only that the developing coefficient a∈(-1/λ,1/1-λ) which exceeds the traditional region a ∈ (- 2,2) , but also the self-adaptive GM (1,1, λ) model coefficient a ∈ ( A is indeed applicable to the high-growth series, and has a higher fitting and prediction precision than the conventional GM (1,1) model.