基于PCA-Logit模型的无车承运人风险预警

发布时间：2018-01-24 14:54

本文关键词： 无车承运人道路运输风险预警 PCA-Logit模型最大似然估计　出处：《江苏大学学报(自然科学版)》2017年03期 　论文类型：期刊论文

【摘要】：为了评估无车承运人承运经营的风险,提高风险预警的准确性,提出了PCA-Logit无车承运人风险预警模型.根据无车承运人经营过程中风险的主要来源,构建了包含26项指标的风险预警指标体系,在对指标体系进行主成分分析(PCA)的基础上,通过Logit风险预警模型计算无车承运人的风险概率.选取15家无车承运人企业进行案例分析,提取了累计方差贡献率达91.03%的8个主成分,根据因子载荷矩阵值得出8个主成分的表达式,采用最大似然估计法对Logit风险预警模型进行参数标定,并计算得到15家无车承运人企业的风险预警值.结果表明:无车承运人企业的风险概率与企业的管理水平、运输市场的市场竞争强度、托运人及实际承运人谈判价格的优势和托运货物的属性等因素具有较强的相关性.
[Abstract]:In order to evaluate the risk of non-vehicle carrier carrying operation, improve the accuracy of risk warning. This paper puts forward the risk warning model of PCA-Logit carrier without vehicle. According to the main source of risk in the course of operation, a risk warning index system including 26 indexes is constructed. On the basis of the principal component analysis (PCA) of the index system, the risk probability of the carrier without vehicle is calculated by using the Logit risk early warning model, and 15 non-vehicle carrier enterprises are selected to carry out the case study. Eight principal components with cumulative variance contribution rate of 91.03% were extracted, and the expression of 8 principal components was given according to the factor load matrix. The maximum likelihood estimation method is used to calibrate the risk early warning model of Logit. The results show that the risk probability, the management level and the market competition intensity of the non-vehicle carrier enterprises are obtained by calculating the risk early warning value of 15 non-vehicle carrier enterprises. The advantage of negotiation price between shipper and actual carrier has strong correlation with the attribute of consignment.
【作者单位】：长安大学汽车学院;交通运输部管理干部学院;辽宁省交通高等专科学校;
【基金】：国家自然科学基金资助项目(51278057) 交通运输部应用基础研究项目(2014319226270) 辽宁省交通高等专科学校优秀人才支持项目(2015lnccrcky06)
【分类号】：U491
【正文快照】： 引文格式:常连玉,胡大伟,陈海燕,等.基于PCA-Logit模型的无车承运人风险预警[J].江苏大学学报(自然科学版),2017,38(3):273-279.无车承运人是一种区别于运输企业、货运代理企业等的一种企业形式,通过整合运输市场资源,可以更好地满足货主(托运人)的需求,同时可以为运输企业争

【参考文献】