摘要
针对现有非等间距GM(1,1)模型存在的不足,借鉴分段线性插值将非等间距序列等间距化的思想,以非等间距的方法建立了一种新的非等间距GM(1,1)模型,模型不需计算插值数据,可直接利用原始数据建模.然后通过赋予原始数据下标序列变换系数,利用平均模拟相对误差最小的原则确定各参数,建立优化后的非等间距GM(1,1)模型.最后通过算例测试和应用实例表明提出模型的有效性和可行性.
Aiming at the limitations of the existing non-equidistant GM （1,1） model and us- ing the ideas of equal spacing to non-equidistant sequence by piecewise linear interpolation, a new non-equidistant GM （1, 1） model is established with non-equidistant method. The model doesn＇t need to calculate the interpolation data and can directly use the original data model- ing. With transforming the original data subscript sequence and determining the parameters by the principle of the minimum average relative error, the optimized non-equidistant GM （1,1） model is established; Finally, taking a numerical example and an application example to show the effectiveness and feasibility of the proposed model.
出处
《数学的实践与认识》
北大核心
2017年第4期251-257,共7页
Mathematics in Practice and Theory
基金
广东普通高校国际暨港澳台合作创新平台及国际合作重大项目(2015KGJHZ023)
