基于偏最小二乘回归的铁路运输收入分析研究
|
摘要
铁路运输收入是反映铁路运输企业盈利能力的重要指标之一,分析影响铁路运输收入的关键因素,建立计算铁路运输收入的数学模型,对提高计算速度,促进铁路运输收入增长具有重要意义。本文基于中国铁道年鉴统计数据,选取铁路营业里程、车辆数、旅客与货物的发送量、周转量等9项影响因素,采用偏最小二乘回归模型解决因素间多重相关性的问题,并与最小二乘回归模型结果进行对比。研究结果表明:运用偏最小二乘回归法计算结果精确可靠,与铁路运输收入呈正相关的主要因素为铁路货车辆数、旅客发送量以及旅客周转量,铁路客车辆数、货物周转量对铁路运输收入影响呈负相关。
Railway transport revenue is one of the most important indicators to reflect the profitability of railway transport enterprises. Analyzing the key factors affecting the railway transport revenue and establishing the mathematical model for calculating the railway transport revenue are of great significance to speeding up the calculation and increasing the railway transport revenue. Based on the statistics from China Railway Yearbook, 9 influential factors such as railway mileage, number of vehicles, delivery volume of passengers and freight, turnover rate and so on were selected. Partial least-squares regression model was used to solve the problem of multiple correlations between factors, compared with the results using the least square regression model.The results show that the method of partial least squares regression is accurate and reliable. The main factors positively related to the revenue from railway transportation are the number of railway vehicles, the volume of passengers sent and the turnover of passengers. However, the influence of railway passenger cars and freight turnover on railway transport revenue is negatively correlated.
Railway transport revenue is one of the most important indicators to reflect the profitability of railway transport enterprises. Analyzing the key factors affecting the railway transport revenue and establishing the mathematical model for calculating the railway transport revenue are of great significance to speeding up the calculation and increasing the railway transport revenue. Based on the statistics from China Railway Yearbook, 9 influential factors such as railway mileage, number of vehicles, delivery volume of passengers and freight, turnover rate and so on were selected. Partial least-squares regression model was used to solve the problem of multiple correlations between factors, compared with the results using the least square regression model.The results show that the method of partial least squares regression is accurate and reliable. The main factors positively related to the revenue from railway transportation are the number of railway vehicles, the volume of passengers sent and the turnover of passengers. However, the influence of railway passenger cars and freight turnover on railway transport revenue is negatively correlated.
引文
[1]曹庆山.铁路运输与社会经济适应性分析[D].成都:西南交通大学,2010.
[2]傅选义.铁路运输经济的若干问题[J].北京交通大学学报(社会科学版),2016,15(01):1-10.
[3]毕守锋.铁路运营经济效益指标及其关联分析研究[D].北京:北京交通大学,2012.
[4]《中国铁道年鉴》编辑部.中国铁道年鉴[M].北京:北京中国铁道出版社,2007-2013.
[5]S Wold, A Ruhe, H Wold, W J Dunn. The Collinearity problem in linear regression,the partial least squares(PLS)approach to generalized inverses[J]. SIAM Journal on Scientific and Statistical Co-puting(S0196-5204), 1984, 5(3):735-743.
[6]薛锋,吕丹,罗建.铁路交通基础设施对区域经济发展的支撑力研究——基于西部地区面板数据[J].综合运输,2018,40(01):5-10.
[7]王如义,王慈光,郭孜政,唐建桥.铁路货物周转量的半参数回归模型预测[J].西南交通大学学报,2008(01):96-100.
[8]于松青,林盛.基于偏最小二乘回归的山东省电力需求预测分析[J].干旱区资源与环境,2015,29(02):14-20.
[9]陈秋玲,吴干俊.长三角城市安全生产经济社会风险因子分析——基于偏最小二乘回归[J].中国安全科学学报,2012,22(10):8-13.
[10]王攀.偏最小二乘回归与核偏最小二乘回归的应用[D].昆明:昆明理工大学,2014.
[11]朱信山,周智勇.基于偏最小二乘神经网络的公路通道交通量预测组合模型[J].公路,2007(07):157-160.
[12]陆洪涛.偏最小二乘回归数学模型及其算法研究[D].北京:华北电力大学,2014.
[13]毛李帆,江岳春,龙瑞华,李妮,黄慧,黄珊.基于偏最小二乘回归分析的中长期电力负荷预测[J].电网技术,2008(19):71-77.
[2]傅选义.铁路运输经济的若干问题[J].北京交通大学学报(社会科学版),2016,15(01):1-10.
[3]毕守锋.铁路运营经济效益指标及其关联分析研究[D].北京:北京交通大学,2012.
[4]《中国铁道年鉴》编辑部.中国铁道年鉴[M].北京:北京中国铁道出版社,2007-2013.
[5]S Wold, A Ruhe, H Wold, W J Dunn. The Collinearity problem in linear regression,the partial least squares(PLS)approach to generalized inverses[J]. SIAM Journal on Scientific and Statistical Co-puting(S0196-5204), 1984, 5(3):735-743.
[6]薛锋,吕丹,罗建.铁路交通基础设施对区域经济发展的支撑力研究——基于西部地区面板数据[J].综合运输,2018,40(01):5-10.
[7]王如义,王慈光,郭孜政,唐建桥.铁路货物周转量的半参数回归模型预测[J].西南交通大学学报,2008(01):96-100.
[8]于松青,林盛.基于偏最小二乘回归的山东省电力需求预测分析[J].干旱区资源与环境,2015,29(02):14-20.
[9]陈秋玲,吴干俊.长三角城市安全生产经济社会风险因子分析——基于偏最小二乘回归[J].中国安全科学学报,2012,22(10):8-13.
[10]王攀.偏最小二乘回归与核偏最小二乘回归的应用[D].昆明:昆明理工大学,2014.
[11]朱信山,周智勇.基于偏最小二乘神经网络的公路通道交通量预测组合模型[J].公路,2007(07):157-160.
[12]陆洪涛.偏最小二乘回归数学模型及其算法研究[D].北京:华北电力大学,2014.
[13]毛李帆,江岳春,龙瑞华,李妮,黄慧,黄珊.基于偏最小二乘回归分析的中长期电力负荷预测[J].电网技术,2008(19):71-77.