基于多次修正残差灰色模型的铁路客流预测  

Prediction of Railway Passenger Flow Based on Residual Error Corrections of the Gray Model

作　　者：骆晨[1] 刘澜[1] 

机构地区：[1]西南交通大学交通运输与物流学院,四川成都610031

出　　处：《华东交通大学学报》2013年第4期19-23,共5页Journal of East China Jiaotong University

摘　　要：为了提高全国铁路客流预测精度,针对铁路客流变化的影响因素及特点,对GM(1,1)模型进行改进,提出了多次修正残差灰色模型的铁路客流预测方法。克服了传统的GM(1,1)模型的指数函数预测精度较差、单次修正残差精度不够的缺点,通过多次残差修正,减少预测误差。结合2003—2010年铁路客运数据实例分析。结果证明该方法预测误差小、精度高、计算简便、可操作性强,为铁路客流预测提供了一种更为可行的途径。In order to increase the forecast accuracy of the national railway passenger flow,this study improves GM（1,1）model and proposes the railway passenger flow prediction method based on residual error gray model in light of factors and characteristics of railway passenger flow.By conquering some weaknesses,such as the inaccurate prediction of the exponential function in traditional GM（1,1）model and the inaccurate single residual correction,the forecast error is reduced with many times of residual modification.Based on the analysis of railway passenger flow data during 2003—2010,the prediction method based on residual error gray model has such advantages as small error,high precision,easy calculation and operability,which provides a feasible approach for the railway passenger flow forecast.

关 键 词：灰色理论 修正残差灰色 客流预测 精度 

分 类 号：U491.14[交通运输工程—交通运输规划与管理]