摘要
通过引入模糊理论中的Vague值(集)相似度量,提出了建立组合预测模型目标函数的两种准则,从而构建了基于新的相关性指标的组合预测模型.运用实例对该方法进行验证,发现基于Vague值(集)相似度所建立的组合预测模型在误差平方和、均方误差、平均绝对误差、平均绝对百分比误差和均方百分误差等方面均比单项预测方法小,表明了所建立组合预测模型的有效性.
In this paper, through introducing the concept of similarity measures between vague values or sets in the theory of fuzzy, we put forward two kinds of criterion as the objective function to build the combination forecast model. This model is based on the new correlation index. Then we use an example to verify the validity of the model, and it indicates that the model based on the Vague values （set） similarity is satisfied all the prediction effect evaluation index. The error sum of squares, mean square error, mean absolute error, mean absolute percentage error and mean square percentage error of the model is smaller than single prediction methods, show the effectiveness of the combination forecast model is set up.
出处
《合肥学院学报（自然科学版）》
2015年第3期8-12,共5页
Journal of Hefei University :Natural Sciences
基金
国家社科基金青年项目"优性区间型组合预测模型构建及其有效性理论的研究"(13CTJ006)资助
关键词
组合预测
Vague相似度
目标函数
评价准则
combination forecast
Vague similarity measure
objective function
evaluation criteria
