摘要
针对传统组合预测模型中各单项预测模型赋权问题的不足,提出一种基于Theil不等系数的IOWA算子最优变形组合预测模型。利用IOWA算子,依据各单项预测模型预测值各时间点精度的不同,赋予不同的权系数,建立最优组合预测模型,并将其应用于变形数据分析与预测。实验结果表明,本文方法的预测结果比各单项模型与传统组合模型的结果更准确。
In this paper, which is concerned with the weighting problem of individual forecasting mod- els in the traditional combination forecasting model, we propose an optimal deformation combination forecasting model based on the Theil inequality coefficient and the IOWA operator. We use the IOWA operator to endow different weight coefficients based on the different accuracy of each time with re- spect to the predicted value of each individual forecasting model, and then establish the optimal combi- nation forecasting model. The model is applied in deformation data analysis and prediction. The ex- perimental results show that the forecast results of the proposed approach are more accurate than those of individual models and the traditional combination model.
出处
《大地测量与地球动力学》
CSCD
北大核心
2018年第1期83-86,共4页
Journal of Geodesy and Geodynamics
基金
国家自然科学基金(41461089)
湖南省科技计划(2016JJ6018)
湖南省教育厅科技计划(16B046)
益阳市科技计划(2016C006)~~
关键词
Theil不等系数
10WA算子
组合预测模型
变形分析
权系数
Theil inequality coefficient
IOWA operator
combination forecasting model
deformation analysis
weight coefficient