摘要
为对区间模糊数进行研究分析,本文首先将区间模糊数转换为等量信息的精确数序列,结合ARIMA模型和BP神经网络的优点,对转换后的精确数序列分别建立以ARIMA模型与BP神经网络的组合模型,最后区间模糊数序列的拟合值和预测值可利用还原公式得到。文章以二元区间模糊数为研究对象,建立了以ARIMA模型与BP神经网络的组合模型。数值实验表明,该方法有效可行,预测精度要优于ARIMA模型。
Through the research and analysis of interval fuzzy numbers,first,the interval fuzzy numbers are converted into exact sequence of equal information.Then,ARIMA model with BP neural network are combined.The advantage of this method is that the combination model of ARIMA model and BP neural network is established for the converted precise sequence of numbers.Finally,the fitting value and predicted value of interval fuzzy sequence can be obtained by using the reduction formula.This paper takes the binary interval fuzzy number as the research object,and establishes the combination model of ARIMA model and BP neural network through numerical value.Experiments show that the method is effective and feasible,and the prediction accuracy is better than ARIMA model.
作者
谢小军
马虹
杨付贵
邱云兰
XIE Xiaojun;MAHong;YANG Fugui;QIU Yunlan(Department of Basic Teaching,Guangzhou College of Technology and Business,Guangzhou Guangdong 510850,China;School of Financial Mathematics and Statistics,Guangdong University of Finance,Guangzhou Guangdong 510521,China)
出处
《阜阳师范大学学报（自然科学版）》
2020年第3期9-13,18,共6页
Journal of Fuyang Normal University:Natural Science
基金
广州工商学院科研课题立项项目(KA201933)资助。
关键词
时间序列
区间模糊数
预测
组合模型
time series
interval fuzzy number
prediction
combination model
