This paper is an attempt to better understand the endogenous relationships between urban form,public transit accessibility, and daily travel distance. A model of two simultaneous equations isimplemented, taking into account interaction between ownership of vehicles and choice ofhousehold location as explanatory endogenous variables for total distance traveled byrespondents. Choice of household location is defined based on cluster analysis (neighborhoodtypology) driven by land use mix, population density and transit accessibility. Controlling forsocio-economic variables, the impacts of neighborhood typologies combined with car ownershiplevels as endogenous choices are estimated using a simultaneous equation model. This researchmakes use of the Quebec City origin/destination survey data from 2001, which include more than50,000 individual respondents. Among other results, it was confirmed the presence ofendogeneity. Without taking this into account, it is found that the joint effects of car ownershipand household location choices are underestimated. Based on the simultaneous model, totaldistance traveled by individuals is primarily influenced by employment status as well ashousehold structure. In fact, total distance per individual has an average growth of 50% whenworking full time. This also increases in 5.7% per child and decreases in 2.4% per person(economy of scale). While urban form and transit supply variables introduced individually intothe model have little influence on distance traveled, neighborhood type as an endogenousvariable has more explanatory power. For example, an individual living in the periphery andowning a vehicle that moves downtown and abandons his or her vehicle reduces their total tripdistance by nearly 75%.
展开▼
