突发事件对我国铁路客运量的干预特征研究

基于新冠肺炎疫情等突发事件对人们日常生活出行的影响,结合X-13ARIMASEATS季节调整模型的自动识别最优ARIMA模型和检测突发事件离群值功能,使用脉冲函数和阶梯函数设计基于离群值的突发事件的干预变量,构建铁路客运量的时间序列ARIMAX干预模型,对铁路客运量近年受到的SARS疫情、铁路客票实名制政策和新冠肺炎疫情等突发事件的冲击趋势进行干预比较分析。结果显示,SARS和新冠肺炎疫情对铁路客运量冲击较大,SARS疫情在冲击滞后的第5~6期铁路客运量基本得到恢复,新冠肺炎疫情对铁路客运量冲击一直在持续中,铁路客运实名制政策实施社会性较强,冲击具有波动性和不稳定性特征,持续时间较短;相对季节调整模型的趋势分析优势,干预模型拟合预测精度显著高于季节调整模型,预测显示我国铁路客运量在缓慢持续回暖中。

Research on the Intervention Characteristics of Emergencies to Railway Passenger Volume in China

In view of the impacts of emergencies such as the COVID-19 pandemic on people’s daily travel and life,this paper leveraged the functions of the X-13ARIMA-SEATS seasonal adjustment model to automatically identify the optimal autoregressive integrated moving average(ARIMA)model and detect outliers in emergencies.The impulse function and the step function were used to design intervention variables of emergencies based on outliers.This paper constructed a time-series ARIMAX intervention model for railway passenger volume and compared the trends in the impacts of emergencies such as the SARS epidemic,the real-name policy for railway passenger tickets,and the COVID-19 pandemic on the railway passenger volume in recent years.The results show that SARS and the COVID-19 pandemic have great impacts on railway passenger volume.The volume recovered in the fifth to sixth phases after the SARS epidemic.The impact of the COVID-19 pandemic on railway passenger volumeis continuing.The implementation of the real-name policy for railway passenger transportation was society-wide,and the impact was characterized by volatility,instability,and short duration.In spite of the trend analysis advantage of the seasonal adjustment model,the fitting and prediction accuracy of the intervention model is significantly higher than that of the seasonal adjustment model.The prediction shows that the railway passenger volume in China is in a slow and continuous recovery.