摘要
灰色预测模型背景值是影响灰色模型性能的重要参数,传统基于紧邻均值的背景值构造方式,容易导致背景值大小受到建模序列中极端值的影响。文章提出了一种基于三参数背景值构造的新方法,该方法提高了灰色预测模型背景值的平滑效果,弱化了建模序列中极端数据对灰色预测模型性能的影响。构建了一种新型灰色预测模型NGM3（1,1）,通过案例比较分析,验证了NGM3（1,1）模型性能优于传统的灰色预测模型。
The background value of grey prediction model is an important parameter that influences the performance of grey model. The traditional construction method based on the background value adjacent to the mean is easy to cause the background value size to be affected by the extreme value in the modeling sequence. This paper proposes a new method based on background value of three parameters, which improves the smoothing effect of the background value of the grey prediction model, and weakens the influence of the extreme data on the performance of the grey prediction model in the modeling sequence. And then the paper constructs a new grey prediction model NGM3（1,1）. Finally through the case comparison analysis, the paper verifies that the performance of the proposed model is better than that of the traditional grey prediction model.
作者
杨孝良
周猛
曾波
Yang Xiaohang;Zhou Meng;Zeng Bo(Center of Economic Research on Upper Reaches of Yangtze River;National Research Base of Intelligent Manufacturing Service,Chongqing Technology and Business University,Chongqing 400067,China)
出处
《统计与决策》
CSSCI
北大核心
2018年第19期14-18,共5页
Statistics & Decision
基金
中国科协重大招标项目(2016ZCYJ06)
重庆市社科规划委托项目(2016WT37)
