Empirical approaches to combining revealed and stated preference data: Some recent developments with reference to urban mode choice

This paper replaces the nested logit ‘trick’ used in estimation of a modal choice model, when stated preference (SP) and revealed preference (RP) data are pooled, with an error component mixed logit model that can (i) relax the IID condition attached to subsets of alternatives in nested logit models, (ii) accommodate correlated observations, and (iii) reveal the relevant scale parameter for subsets of alternatives. Such a model can also incorporate state dependence between data types and preference heterogeneity on observed attributes. An example of choice among bus, train and car modes illustrates the difference in empirical evidence, especially elasticity outputs.