摘要
针对传统回归模型需要的数据量大且建模复杂等缺陷,提出了一种基于灰色关联度和GM(1,1)的灰色组合预测模型;从灰色关联度的视角寻找数据之间的依赖关系,运用GM(1,1)模型预测数据关系的未来发展趋势,进而建立因变量的预测模型;模型体现了回归分析基于事物因果关系的建模思想,同时又具有灰色理论小样本建模的特点;应用该模型对我国2007和2008年度的GDP进行预测,预测结果表明了该模型的有效性及实用性。
Traditional regression models need enormous historic data, and it is very hard to build this kind of model. With those limitations, a grey combined forecast model, based on grey incidence degree and GM （1,1） model, is put forward. This model seeks dependency relationships of sequences based on the grey incidences degree, employs the GM （1,1） model to forecast the trend of development among data, and then a forecast model about dependent variable is set up. The model not only reflects the modeling thought of regression analysis based on causality, but also has the characteristics of small example for building model. The GDP in 2007 and 2008 are forecasted with this model,and the results illustrate the validity and practicability of the novel model.
出处
《中国管理科学》
CSSCI
北大核心
2009年第5期150-155,共6页
Chinese Journal of Management Science
基金
国家自然科学基金资助项目(70701017
709010141
70971064
90924022)
教育部博士点科研基金(200802870020)
江苏高等学校优秀创新团队(P0702)
南京航空航天大学创新群体(Y0553)
特聘教授科研创新基金(1009-260812)
关键词
预测技术
组合预测模型
灰色关联度
回归分析
GM(1
1)模型
forecast technology
combined forecast model grey incidence degree
regression analysis
GM （1, 1） model
