摘要
针对诱导有序加权平均(induced ordered weighted averaging,IOWA)组合模型中,单项预测方法精度相同时诱导值可能存在多种不同排列顺序的问题,提出综合考虑预测方法的历史性能确定其唯一排序;针对实践中单项预测方法较多时,IOWA组合模型构建过程十分复杂的问题,通过构建适合于Matlab软件计算的最优化模型矩阵形式,降低计算复杂性;结合舰船装备维修费预测中样本小、预测方法多的特征,采用各单项预测方法的历史平均拟合精度来估计其预测期诱导值的排序,并以此计算组合预测值。最后实例说明了该方法的有效性。
In the induced ordered weighted averaging ferent ordinal arranges of induced values when individual （IOWA） combination model, there may t）e many difforecasting methods have the same precision. Aiming at this problem, the unique ordering can be determinated by taking the history performance of each forecasting method into account synthetically. Meanwhile, the constructing process of the IOWA combination model is very complex when there are many individual forecasting methods. Then the matrix form of the optimization model that is fit for operating the Matlab software is described, which can reduce the computational complexity. Considering the characters of small samples and many forecasting methods in forecasting ship equipment maintenance cost, the ordering of induced values in forecasting period is estimated by the historical averaging fitting precision of individual forecasting methods, and the value of combination forecast is calculated. Finally an example is given to show the availability of this method.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2012年第6期1176-1181,共6页
Systems Engineering and Electronics
基金
海军工程大学基金(HGDQNJJ041
HGDQNEQJJ11016)资助课题
关键词
诱导有序加权平均
组合预测
小样本
舰船装备
维修费
induced ordered weighted averaging （IOWA）
combined forecasting
small sample
ship equipment
maintenance cost
