A JOINT TOUR-BASED MODEL OF TOUR COMPLEXITY, PASSENGER ACCOMPANIMENT, VEHICLE TYPE CHOICE, AND TOUR LENGTH

摘要

Tour-based model systems are increasingly being deployed to microsimulate daily activity-travelpatterns of individuals. There are a host of tour attributes of interest that are modeled withinthese systems. However, a dimension that is often missed is that of vehicle type choice, avariable of considerable importance in the energy consumption and emissions estimation arena.Another issue that arises is that most tour attributes are modeled independently or sequentiallywith loose coupling across the models, thus ignoring important endogeneity effects that mayexist across multiple tour dimensions. This paper considers four key dimensions of tours – tourcomplexity, passenger accompaniment, vehicle type choice, and tour length – with a view todeveloping a joint simultaneous equations model system of tour choices while accounting for thepresence of correlated unobserved attributes affecting multiple dimensions through appropriateerror covariance structures. The paper makes an important methodological contribution bydescribing and formulating a multi-dimensional joint choice model system capable ofaccommodating a variety of endogenous variable types (discrete and continuous). The papermakes an important empirical contribution by providing evidence on the nature of therelationships among these tour dimensions of interest within the context of a joint model. Themodel system is estimated on a sample of tours from the 2009 National Household TravelSurvey of the United States. In general, it is found that there is significant evidence of correlatedunobserved factors across these tour dimensions and that vehicle type choice affects tour length,a finding with potentially important policy implications.