基于GM(1,1)季节指数模型的快递业务量预测  

Prediction of express business volume based onseasonal index model

作　　者：王芳 叶耀军[1] WANG Fang;YE Yaojun(School of Sciences,Zhejiang University of Science and Technology,Hangzhou 310023,Zhejiang,China)

机构地区：[1]浙江科技学院理学院,杭州310023

出　　处：《浙江科技学院学报》2021年第6期504-508,526,共6页Journal of Zhejiang University of Science and Technology

基　　金：浙江省自然科学基金项目(LY17A010009);南京航空航天大学灰色系统主题创新区项目(GS2019012)。

摘　　要：快递业务量存在季节性波动和总体变化等问题,为了更好地预测其趋势,提出基于季节指数的灰色系统模型预测方法。首先以2014—2019年的全国快递季度业务量为基础,分别建立GM(1,1)模型和GM(1,1)季节指数模型,并计算出2014—2019年的季度快递量拟合值;然后与实际值进行比较,同时做出拟合效果图;最后经过模型的精度指标计算,比较2种方法的模拟效果。结果显示,GM(1,1)季节指数模型精度指标平均绝对误差、均方差和平均绝对百分误差,分别降低了36.840%,74.547%和50.314%。这表明该模型模拟出来的数值更贴近实际值,不仅精度高,而且在数据变化上体现季节波动性,更适合快递业务量的预测。There are seasonal fluctuations and overall changes in express business volume.In order to better predict its trend,a grey system model prediction method was proposed on the basis of seasonal index.Firstly,based on the quarterly express business volume of China from 2014 to 2019,a grey model and a seasonal index model were established respectively,calculating the fitting value of the quarterly express business volume from 2014 to 2019 to be compared with the actual value and then drawing up the fitting effect diagram.Finally,the simulation effects of the two methods were compared through the precision index calculation of the models.The results show that the mean absolute error,the mean square error and the mean absolute percentage error of precision index of the seasonal index model are reduced by 36.840%,74.547% and 50.314% respectively.This shows that the numerical value simulated by this model is closer to the actual value,not only with high precision,but also with seasonal fluctuations in data changes,which is more suitable for predicting the express business volume.

关 键 词：季节指数模型 GM(1 1)模型 快递业务量预测 

分 类 号：F259.2[经济管理—国民经济]