摘要
为给铁路运输部门规划设计提供科学准确的短期铁路预测客运量,以2005—2018年铁路月客运量为基础,根据其增长趋势和周期性变化规律,分别采用季节性指数平滑法和季节差分自回归移动平均法(seasonal autoregressive integrated moving average,SARIMA)建立模型,预测2019年铁路客运量,并与实际客运量对比。以平均绝对百分比误差(root mean square error,MAPE)和均方根误差(root mean square error,RMSE)为衡量标准,对比分析2种方法的预测结果。分析结果表明:与指数平滑法相比,应用SARIMA模型使预测的铁路客运量的MAPE减少56.26%,RMSE减少64.61%,SARIMA模型更适合对铁路客运量进行短期预测,精度较高。
In order to provide the scientific and accurate short-term railway forecasting passenger volume for the planning and design of the railway transportation departments,based on basic data of monthly railway passenger volume from 2005 to 2018,and according to its growth trend and periodic change law,the seasonal exponential smoothing method and seasonal autoregressive integrated movement average(SARIMA)are used to establish a model to predict the railway passenger volume in 2019 and compare with the actual passenger volume.The prediction results of the two methods are compared by means of the mean absolute percentage error(MAPE)and root mean square error(RMSE).The analytical results show that compared with the exponential smoothing method,the use of SARIMA model can make MAPE of the forecasting railway passenger volume reduce by 56.26%and RMSE by 64.61%,and the SARIMA model is more suitable for the short-term prediction of railway passenger volume with higher precision.
作者
王雷
金勇
刘岩
WANG Lei;JIN Yong;LIU Yan(School of Transportation Engineering, Dalian Jiaotong University, Dalian 116028, China;Dalian Land Space Planning and Design Co. , Ltd. , Dalian 116000,China)
出处
《山东交通学院学报》
CAS
2020年第3期25-32,47,共9页
Journal of Shandong Jiaotong University
基金
辽宁省教育厅基金(JDL2017010)
辽宁省自然科学基金(20180550262)
辽宁省博士科研启动基金(20170520210)。
关键词
铁路客运量
指数平滑法
季节差分移动自回归模型
预测
railway passenger volume
exponential smoothing method
seasonal autoregressive integrated moving average
prediction
