摘要
机场旅客吞吐量的影响因子通常存在滞后期,论文在选定了影响因素之后,通过随机森林重要度指数,确定各因子的滞后期;接着,以1987-2018年南京禄口国际机场的数据建立吞吐量预测的逐步回归模型;最后,对模型效果进行评价并基于建立的模型对2019年和2020年南京禄口国际机场的吞吐量进行预测。结果表明:影响因子对旅客吞吐量的影响存在1~9年不等的滞后期;通过与未考虑滞后期的模型进行对比,发现考虑滞后期的模型拟合效果更优,MAE减小显著。因此,将滞后期引入机场旅客吞吐量预测是很有必要的。
The impact factor of airport passenger throughput usually has a lag period. After selecting the influencing factors, the paper determines the lag period of each factor by the random forest importance index. Then, a stepwise regression model for throughput prediction is established using the data of Nanjing Lukou International Airport from 1987 to 2018. Finally, the model effect is evaluated and the throughput of Nanjing Lukou International Airport in 2019 and 2020 is predicted based on the established model. The results show that the influence of impact factors on passenger throughput has a lag period ranging from 1 to 9 years. By comparing with the model without considering the lag period, it is found that the model fitting effect considering the lag period is better and the reduction is significant. Therefore, it is necessary to introduce the lag period into the airport passenger throughput forecast.
作者
刘月
朱金福
陈娴
Liu Yue;Zhu Jinfu;Chen Xian(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)
出处
《华东交通大学学报》
2019年第6期64-69,共6页
Journal of East China Jiaotong University
基金
南京航空航天大学研究生创新基地(实验室)开放基金项目(kfjj20180723)
关键词
滞后期
预测
随机森林
逐步回归
神经网络
lag period
prediction
random forest
stepwise regression
neural network
