THE ESTIMATION OF MULTINOMIAL LOGIT MODELS OF JOINT LOCATION AND TRAVEL MODE CHOICE FROM AGGREGATED DATA*

Exploring the influencing factors of intercity travel mode choice can reveal passengers’ travel decision mechanisms and help traffic departments to develop an effective demand management policy. To investigate these factors, a survey was conducted in Xi’an, China, to collect data about passengers’ travel chains, including airplane, high-speed railway (HSR), train, and express bus. A Bayesian mixed multinomial logit model is developed to identify significant factors and explicate unobserved heterogeneity across observations. The effect of significant factors on intercity travel mode choice is quantitatively assessed by the odds ratio (OR) technique. The results show that the Bayesian mixed multinomial logit model outperforms the traditional Bayesian multinomial logit model, indicating that accommodating the unobserved heterogeneity across observations can improve the model fit. The model estimation results show that ticket purchasing method, comfort, punctuality, and access time are random parameters that have heterogeneous effects on intercity travel mode choice.

Download Full-text

Machine Learning Travel Mode Choices: Comparing the Performance of an Extreme Gradient Boosting Model with a Multinomial Logit Model

Transportation Research Record Journal of the Transportation Research Board ◽

10.1177/0361198118773556 ◽

2018 ◽

Vol 2672 (47) ◽

pp. 35-45 ◽

Cited By ~ 9

Author(s):

Fangru Wang ◽

Catherine L. Ross

Keyword(s):

Machine Learning ◽

Mode Choice ◽

Multinomial Logit ◽

Choice Modeling ◽

Gradient Boosting ◽

Travel Mode ◽

Extreme Gradient Boosting ◽

Travel Mode Choice ◽

Mnl Model ◽

Mode Choice Modeling

The multinomial logit (MNL) model and its variations have been dominating the travel mode choice modeling field for decades. Advantages of the MNL model include its elegant closed-form mathematical structure and its interpretable model estimation results based on random utility theory, while its main limitation is the strict statistical assumptions. Recent computational advancement has allowed easier application of machine learning models to travel behavior analysis, though research in this field is not thorough or conclusive. In this paper, we explore the application of the extreme gradient boosting (XGB) model to travel mode choice modeling and compare the result with an MNL model, using the Delaware Valley 2012 regional household travel survey data. The XGB model is an ensemble method based on the decision-tree algorithm and it has recently received a great deal of attention and use because of its high machine learning performance. The modeling and predicting results of the XGB model and the MNL model are compared by examining their multi-class predictive errors. We found that the XGB model has overall higher prediction accuracy than the MNL model especially when the dataset is not extremely unbalanced. The MNL model has great explanatory power and it also displays strong consistency between training and testing errors. Multiple trip characteristics, socio-demographic traits, and built-environment variables are found to be significantly associated with people’s mode choices in the region, but mode-specific travel time is found to be the most determinant factor for mode choice.

Download Full-text