高思凡.基于时变权重的区间时间序列组合预测模型构造[J].重庆工商大学学报(自然科学版),2021,38(2):40-47
GAO Si-fan.Construction of Interval Time Series Combined Prediction Model Based on Time varying Weight[J].Journal of Chongqing Technology and Business University(Natural Science Edition),2021,38(2):40-47
基于时变权重的区间时间序列组合预测模型构造
Construction of Interval Time Series Combined Prediction Model Based on Time varying Weight
  
DOI:
中文关键词:  时变权重  区间值  组合预测
英文关键词:time varying weight  interval value  combined prediction
基金项目:
作者单位
高思凡 安徽大学 经济学院合肥 230601 
摘要点击次数: 10
全文下载次数: 4
中文摘要:
      时变权重组合预测模型可以有效反映各预测方法在各时刻点上的预测值对组合预测结果的影响,并可以提高预测精度,基于此,针对区间数时间序列构造组合预测模型的问题,提出一类构造区间数时变权重方法;该方法主要是在3种实数的时变权重求解方法基础上构造相应的3种区间数时变权重,并利用所得出的时变权重构造区间时间序列组合预测模型;为验证该模型的准确性,应用具体算例对模型加以实现,并通过区间预测误差度量指标验证该类模型的预测结果;结果发现:基于3种时变权重求解方法下的区间时间序列组合预测模型均优于各单个方法的预测结果,且基于最优化赋权法得到的时变权重区间组合预测结果的精确度较另外两种有所提高。
英文摘要:
      For the current time varying weighted combination prediction model,it can effectively reflect the influence of the prediction value of each prediction method at each time point on the combination prediction result,and it can effectively improve the accuracy of the prediction,as well as based on the problem of constructing the combination prediction model of the interval number time series,a new method for constructing time varying weights of interval numbers is proposed.This method is mainly based on three real time time varying weights solving methods to construct the corresponding three time varying interval weights,and uses the obtained time varying weights to construct interval time series combined prediction models.In order to verify the accuracy of the model,the model is implemented through specific examples and the prediction results of this type of model are measured by interval prediction error metrics.It was found that the interval time series combination prediction models based on three time varying weights solving methods were better than the prediction results of each single method,and the accuracy of the time varying weight interval combination prediction results obtained based on the optimization weighting method was better than the other two.
查看全文  查看/发表评论  下载PDF阅读器
关闭
重庆工商大学学报自然科学版 版权所有
地址:中国 重庆市 南岸区学府大道19号,重庆工商大学学报编辑部 邮编:400067
电话:023-62769495 传真:
您是第2427642位访客
关注微信二维码