Integrating revealed preference and stated response data into a jointly estimated hierarchical mode choice model
Access status:
Open Access
Type
Working PaperAuthor/s
Hensher, David A.Abstract
Revealed preference and stated response data have both contributed to the development of the literature on behavioural travel demand modelling. Until very recently, these two types of data have been independently used in the estimation of a wide variety of discrete choice applications ...
See moreRevealed preference and stated response data have both contributed to the development of the literature on behavioural travel demand modelling. Until very recently, these two types of data have been independently used in the estimation of a wide variety of discrete choice applications in transport. There is growing interest in exploring the view that both types of data have useful information and that their integration will improve the overall explanatory power of choice models. In this paper, we present the theoretical framework for combining the data sources, and specify a model capable of introducing the two data sets with independent choice outcomes. The approach requires the application of a full information maximum likelihood estimation procedure of the hierarchical logit form. We demonstrate the advantages of the dual data strategy by comparing the results with those obtained from models estimated independently with RP and SR data. Data collected as part of the prefeasibility study of the Very Fast Train Project is used to estimate the set of mode choice models.
See less
See moreRevealed preference and stated response data have both contributed to the development of the literature on behavioural travel demand modelling. Until very recently, these two types of data have been independently used in the estimation of a wide variety of discrete choice applications in transport. There is growing interest in exploring the view that both types of data have useful information and that their integration will improve the overall explanatory power of choice models. In this paper, we present the theoretical framework for combining the data sources, and specify a model capable of introducing the two data sets with independent choice outcomes. The approach requires the application of a full information maximum likelihood estimation procedure of the hierarchical logit form. We demonstrate the advantages of the dual data strategy by comparing the results with those obtained from models estimated independently with RP and SR data. Data collected as part of the prefeasibility study of the Very Fast Train Project is used to estimate the set of mode choice models.
See less
Date
1992-10-01Department, Discipline or Centre
ITLSShare