基于相对熵的最优区间型组合预测模型  

Optimal Interval Combination Forecasting Model Based on Relative Entropy

作　　者：袁宏俊 杜康[2] 胡凌云 Yuan Hongjun;Du Kang;Hu Lingyun(School of Statistics,Dongbei University of Finance and Economics,Dalian Liaoning 116025,China;School of Statistics and Applied Mathematics,Anhui University of Finance and Economics,Bengbu Anhui 233030,China;School of Management Science and Engineering,Anhui University of Finance and Economics,Bengbu Anhui 233030,China)

机构地区：[1]东北财经大学统计学院,辽宁大连116025 [2]安徽财经大学统计与应用数学学院,安徽蚌埠233030 [3]安徽财经大学管理科学与工程学院,安徽蚌埠233030

出　　处：《统计与决策》2020年第11期26-31,共6页Statistics & Decision

基　　金：安徽省教育厅人文社会科学重点研究项目(SK2020A0028,SK2018A0431);辽宁省教育厅科学研究重点攻关项目(LN2019Z12);安徽财经大学重点科研基金资助项目(ACKY1713ZDB);安徽财经大学研究生教育教学研究重点项目(cxjhjyzdi1813)。

摘　　要：文章将传统的区间数转化为二元联系数,分别以二元联系数的同部序列和异部序列作为研究对象,选取相对熵作为优化准则,引入偏好系数,建立基于相对熵的最优区间型组合预测模型,并给出该模型是优性组合预测的性质定理。实例分析结果显示,所构建的区间型组合预测模型可以显著地减少预测误差,有效提高预测精度。This paper transforms the traditional interval numbers into binary relation numbers and takes the same part sequence and different part sequence of binary relation numbers as the research objects respectively, with the relative entropy chosen as the optimization criterion, introducing preference coefficient to establish the optimal interval combination prediction model based on relative entropy. Finally, the paper presents the property theorem of optimal combination prediction. Example analysis results show that the interval combination prediction model can significantly reduce the prediction error and improve the prediction accuracy.

关 键 词：区间型组合预测 二元联系数 相对熵 灵敏度分析 

分 类 号：F224[经济管理—国民经济]