A latent class model with attribute cut-offs to analyze modal choice for freight transport

Abstract

We use stated preference data to analyze modal choice for freight transport when information about attribute cut-offs is introduced into the utility specification. Different choice models are estimated to account for the negative effect produced when these threshold values are violated. In order to better understand the heterogeneity in shippers' preferences, a latent class model incorporating cut-offs penalties is estimated. Our results provide evidence of the existence of differentiated classes of individuals regarding both the perception of the main attributes affecting modal choice in the corridor under analysis and penalties imposed when these attributes do not meet acceptable levels of service.