一类日极端温度的变权系数区间型组合预测模型  

A Class of Interval-type Combined Prediction Models with Variable Weighting Coefficients for Daily Temperature Extremes

作　　者：杨烨 袁宏俊[1] 陈倩如 YANG Ye;YUAN Hong-jun;CHEN Qian-ru(School of Statistics and Applied Mathematics/Anhui University of Finance and Economics,Bengbu Anhui,233030;School of Management,Hefei University of Technology,Hefei Anhui,230009)

机构地区：[1]安徽财经大学统计与应用数学学院,安徽蚌埠233030 [2]合肥工业大学管理学院,安徽合肥230009

出　　处：《山西大同大学学报（自然科学版）》2023年第4期49-55,共7页Journal of Shanxi Datong University(Natural Science Edition)

基　　金：安徽省哲学社会科学规划项目[AHSKY2020D42];安徽省高校科研计划重点项目[2022AH050602,2022AH050565];安徽财经大学科研基金重大项目[ACKYA21004];安徽财经大学研究生科研创新基金项目[ACYC2022363]。

摘　　要：为提高日极端温度的预测精确度,提出了一类基于区间数相关度与GIOWHA算子新型区间数组合预测模型。将区间数转化为等价信息的中心和半径,以相关度作为最优准则,构建基于区间数相关度的定权系数区间型组合预测模型。并引入GIOWHA算子,将区间数预测精度作为诱导因子,构建基于区间数相关度与GIOWHA算子的变权系数区间型组合预测模型。并将变权系数区间型组合预测模型应用于日极端温度区间数序列的预测中,实例证明该模型的有效性,并对模型的参数进行了灵敏度分析。To improve the prediction accuracy of daily temperature extremes,a new class of interval number combination prediction models based on interval number correlation and the GIOWHA operator is proposed in this paper.The interval number is transformed into the centre and radius of equivalent information,and the correlation degree is used as the optimality criterion to construct an interval-type combined prediction model with fixed weight coefficients based on the interval number correlation degree.The GIOWHA operator is also introduced,and the interval number prediction accuracy is used as an inducing factor to construct an interval-type combined prediction model with variable weight coefficients based on the interval number correlation and the GIOWHA operator.The model is applied to the prediction of daily extreme temperature interval number series,and the effectiveness of the model is demonstrated by example,and the sensitivity analysis of the parameters of the model is carried out.

关 键 词：日极端温度 组合预测 相关度 GIOWHA算子 灵敏度分析 

分 类 号：F224.0[经济管理—国民经济]