应用多元线性回归模型的铁路客运量预测

Railway Passenger Volume Forecast Based on Multiple Linear Regression Model

以2004—2016年西宁市既有线铁路客运量及其他相关统计数据为基础样本数据,建立西宁市铁路客运量多元线性回归预测模型,预测人口总数、历年生产总值及接待游客总量对西宁既有线铁路客运量的影响程度,采用Eviews软件作为软件平台,对铁路客运量进行短期预测,最后对模型进行检验。结果表明:游客总量对西宁既有线铁路客运量有着显著的正相关关系,而人口总数、生产总值对铁路客运量有着负相关关系。经检验与应用,利用回归模型得到的预测值与实际值相差较小,精度较高,预测效果良好,适合对西宁铁路客运量的短期预测。

Based on the sample data of the passenger traffic of the existing railway lines and other relevant statistics from 2004 to 2016 in Xining City,a multiple linear regression model for the passenger traffic of Xining City was established to predict the total population,historical total value of production and the total number of visitors received in Xining. Regarding the extent of the impact of passenger traffic on the existing railway lines,Eviews software was used as a software platform to make short-term predictions of railway passenger traffic,and the model was finally tested. The results show that the total number of tourists has a significant positive correlation with the passenger volume of Xining-Cable Railway,while the total population and production value have a negative correlation with the passenger volume of railways. After inspection and application,the regression model obtains a small difference between the predicted value and the actual value,with high precision and good prediction effect,and is suitable for the short-term prediction of Xining railway passenger traffic.