摘要
为了进一步研究呈陡峭变化序列的规律,首先给出了GM(1,1|τ,r)模型及其参数的详细证明,然后提出了一种确定参数τ,r的方法,即灰色时滞绝对关联度,最后通过对近年来河南省国际旅游(外汇)收入分析和预测的实例验证了该方法的可行性。结果发现用该方法确定的参数使所得模型的预测精度相比普通GM(1,1)有了较大的提高,为该模型的后续研究奠定了基础。
In order to investigate the regularity of the steep sequence, the model of GM （ 1,11 τ,r） was firstly introduced and the detailed proof of the model parameters was given. Then, a method to determine the T,r was proposed. At last, the feasibility of this method was verified through the analysis and prediction of the international tourism income （foreign exchange income） in Henan Province. The results show that using this method to determine parameters of GM （1,1 τ,r） has been greatly improved the prediction accuracy of the resulting than the ordinary GM （ 1,1 ） prediction model. The research makes a foundation for the further research of GM（ 1,1 1τ,r） at the same time.
出处
《武汉理工大学学报（信息与管理工程版）》
CAS
2013年第4期536-538,569,共4页
Journal of Wuhan University of Technology：Information & Management Engineering
基金
国家自然科学基金资助项目(70971103)
国家教育部高等学校博士点专项科研基金资助项目(20120143110001)
关键词
灰色时滞
灰色模型
灰色绝对关联度
搜索法
grey time - delay
grey model
grey absolute correlation degree
search method
