摘要
对高速铁路旅客市场进行细分是应用收益管理理论的重要环节.基于京沪高铁的客票数据,选取年龄、性别、出行日期、出行距离、购票方式和提前购票时间6类外显变量作为分类指标,采用潜在类别模型进行高速铁路旅客市场的细分.首先将外显变量概率参数化后代入模型进行建模并利用Mplus软件进行模型求解,模型拟合的AIC和BIC等指标表明,当潜在类别为3类时模型具有较好的效果.然后根据模型参数估计结果对所有数据进行潜在聚类分析,分类正确率达到93%左右,表明分类结果合理,3种类别的旅客在提前购票时间、出行距离等方面具有明显的差异.潜在类别模型的引入可以为我国高速铁路收益管理理论研究和实践应用提供参考借鉴.
Passenger market segmentation is an important part of the application of the revenue management theory for high-speed railway.Based on the ticket data of Beijing-Shanghai high-speed railway,the latent class model is used for passenger market segmentation taking age,sex,travel date,travel distance,mean to get ticket and pre-purchase time as classification indicators.Firstly,the model is set up after manifest variables probability becoming parameters and solved by Mplus software,the model fitting indicators such as AIC and BIC show that three potential categories are better.Then all data are clustered according to the parameter estimation results,the classification accuracy can reach about 93%,showing that segmentation result is suitable,three categories of passengers have significant differences in terms of prepurchase time,travel distance and so on.The latent class model can provide a reference for theoretical research and practical application of high-speed railway revenue management in China.
作者
乔珂
赵鹏
文佳星
QIAO Ke ZHAO Peng WEN Jia-xing(School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China)
出处
《交通运输系统工程与信息》
EI
CSCD
北大核心
2017年第2期28-34,共7页
Journal of Transportation Systems Engineering and Information Technology
基金
国家自然科学基金项目(u1434207)
中国铁路总公司科技研究计划项目(2015X006-G)~~
关键词
铁路运输
旅客市场细分
潜在类别模型
高速铁路
条件概率
railway transportation
passenger market segmentation
latent class model
high-speed railway
conditional probability
