基于RP/SP联合数据的非集计模型应用研究

实际调查RP数据(Revealed preference Data)具有可靠性,而意向调查SP数据(Stated preference Data)具有灵活性。为了更好地将两类数据结合起来对新的交通方式进行预测,以长株潭城际轨道交通方式划分为例,介绍了数据的调查方法,利用正交设计法对SP调查表格进行了设计。分别建立基于RP和SP数据的Logit模型,鉴于RP与SP模型中随机项的差异性,通过引入SP比例参数,构造RP/SP联合数据模型,并利用3种模型对未来的交通方式分担量进行了预测,最后对预测结果进行了比较分析。实例结果表明,基于RP/SP联合数据的logit模型能够平衡两类数据之间的误差相互影响,并能够得到更为合理的交通方式划分预测结果。

Research of disaggregate model application based on combined RP/SP data

RP data were more reliable and SP data were more flexible.In order to join two types of data to forecast the new mode,taking the mode split of the Changzhutan intercity railway for example,the investigation method was introduced,and the orthogonal design was applied in the SP questionary.The logit model was established based on RP data and SP data respectively,because of the differences in the random terms between RP model and SP model,through introducing the SP scale parameter,a disaggregate model based on the combination data of RP and SP was established.Using the three types of model,the future share of the traffic mode was forecast and the results were compared and analyzed.Experimental results show that the logit model with combined RP/SP data can easily account for the scaling and unobserved error correlations and provides a more reasonable result.