摘要
为解决基于时间序列数据对象的统计分析问题,提出了一种全新的灰色相关分析方法并定义了其数学模型。灰色相关分析方法充分考虑数据对象随时间的变化而改变的特征,突出了最新的数据在分析结果中的重要作用,使相关性分析对新数据保持高敏感性,尽量弱化时间久远的数据影响,准确地反映了近期各对象间的相关性。利用面向对象程序设计原理,使用面向对象开发工具进行了系统实现,并进行了大量实验测试,实验结果显示,灰色相关分析方法对解决时间序列的分析问题,更加准确、可信。将灰色相关分析方法模型应用到我国环境基本情况的统计中,分析结果可为制定环境保护措施提供有力的科学依据。
In order to solve of the data statistical problem on the basis of time-series, the present paper puts forward a brand-new method called grey correlation analysis and the relevant mathematical model has been defined. The advantage of the new method lies in the fact that it will take full consideration of the data＇s corresponding changeability with time and will put the importance of the newest data in the statistical result to the foregrounding position. The high sensitivity of analysis on relativity towards the newest data will be maintained, which will in turn weaken the influences from the far remote data and accurately reflect the relativity among every number target from the recent period. The system was realized using target-oriented program design principle and corresponding development tools to. Moreover, we have carried out a large number of experiments, which show that the analytical result of time-series can be much more accurate and reliable by the method of grey relation analysis. Finally, it can be said that if we can put the model of grey relation analysis into the statistical analysis of the basic conditions in our country＇s environment, the analytical result will produce scientific references for formulating specific strategies for environmental protection.
出处
《吉林大学学报（信息科学版）》
CAS
2007年第1期84-90,共7页
Journal of Jilin University（Information Science Edition）
基金
国家自然科学基金资助项目(60473042)
关键词
灰系统
多元统计分析
灰相关分析
面向对象程序设计
时间序列
grey system
multi-statistical analysis
grey relation analysis
object oriented program （OOP）
time-series
