基于HW-EEMD-SVM模型的民航客运量预测  

Forecast on Passenger Traffic Volume of Civil Aviation Based on HW-EEMD-SVM Model

作　　者：李轩 周新苗[2] 吴晓峰 Li Xuan;Zhou Xinmiao;Wu Xiaofeng(School of Public Administration,Dongbei University of Finance and Economics,Dalian 116000,China;Business School,Ningbo University,Ningbo 315211,China;School of Economics and Management,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]东北财经大学公共管理学院,辽宁大连116000 [2]宁波大学商学院,浙江宁波315211 [3]北京交通大学经济管理学院,北京100044

出　　处：《数量经济研究》2023年第2期189-204,共16页The Journal of Quantitative Economics

摘　　要：作为国民经济和社会发展的重要行业之一,民用航空业在交通运输领域有着重要的作用。本文在把握中国民航客运量数据特征的条件下,基于集成组合预测的思想,将传统时间序列预测方法与机器学习相结合,利用EEMD方法建立了一个HW-EEMDSVM误差修正组合预测模型。之后,选取2008~2019年的中国民航客运量月度数据作为考察样本进行建模及验证模型预测效果,并将之与HW模型、BP模型、SVM模型、EEMD-BP模型、EEMD-SVM模型、HW-EEMD-BP模型的预测效果进行比较,结果表明本文所选的组合预测模型可以为中国民航客运量波动特征下的数据变化趋势提供较为有效的预测。最后,本文还对疫情发生以来中国民航客运量的变化情况进行了影响分析,对疫情可能造成的冲击进行了量化测算。As one of the important industries of national economic and social development,civil aviation industry has developed rapidly in recent years.After understanding the characteristics of passenger traffic volume data of China’s civil aviation,based on the idea of integrated combination forecast,this paper combines the traditional forecasting method of time series with machine learning,and establishes a HW-EEMD-SVM error correction combination forecasting model,with the use of EEMD method.Then,it selects the monthly passenger traffic volume data of China’s civil aviation from 2008 to 2019 as the inspection sample to construct model and verify the prediction effect of the model,and compares them with that of HW,BP,SVM,EEMD-BP,EEMD-SVM,HW-EEMD-BP models.The results show that the combination forecasting model selected in this paper can provide a more effective prediction of the data change trend under the fluctuation characteristics of China’s civil aviation passenger traffic volume.Finally,it also analyzes the impact of changes in China’s civil aviation passenger traffic volume since the COVID-19 epidemic happens,and quantifies the impact it may cause.

关 键 词：民航客运量 集成经验模态分解 支持向量机 组合预测 

分 类 号：F562.3[经济管理—产业经济]