Abstract
This article presents an econometric modelling framework for a discrete-continuous choice structure. The proposed model ensures random utility maximisation (RUM) approach for both discrete and continuous decisions and explicit correlation between the two choices. The continuous choice is modelled as RUM under budget limitation. The model is applied for joint mode choice and departure time choice modelling. The empirical application considers continuous time scale for departure time under daily 24 h time budget constraint. The empirical model is estimated by using data collected in Toronto, Canada. The estimated parameters reveal many behavioural details of commuters’ mode and departure time choices. The RUM-based discrete-continuous model with the incorporation of time budget constraints for the continuous part is a generic econometric model. This article considers modelling ‘mode and departure time’ choice bundle for empirical application of the model. However, the model can be applied for any other similar choice bundle situation, for example, in the case of activity-based travel demand and land use modelling.
Acknowledgements
The author would like to acknowledge the help of Nick Day for data preparation and Tazul Islam for help in model validation. For discussions and access to the data set, the author is grateful to Professor Eric Miller. This research was partially funded by NSERC (Natural Science and Engineering Research Council of Canada) Discovery Grant and the university start-up grant. The author also acknowledges the comments and suggestions of anonymous referees to update the model and article presentation quality.
Notes
Note
For the empirical application we used the TTS dataset, where the day starts at 4 am. So, for the empirical application of the model presented in this article, the earliest departure time refers to not allocating time at-home after 4 am.