Exploring contextual variations in land use and transport analysis using a probit model with geographical weights

A majority of statistical methods used in the analysis of land use and transportation systems implicitly carry the assumption that relationships are constant across locations or individuals, thus ignoring contextual variation due to geographical or socio-economic heterogeneity. In some cases, where the assumption of constant relationships is questionable, market segmentation procedures are used to study varying relationships. More recently, methodological developments, and a greater awareness of the importance of geography, have led to increasingly sophisticated ways to explore varying relationships in land use and transportation modeling. The objective of this paper is to propose a simple probit model to explore contextual variability in continuous-space. Some conceptual and technical issues are discussed, and an example is presented that reanalyzes land use change using data from California’s BART system. The results of the example suggest that considerable parametric variation exists across geographical space, and thus underlines the relevance of contextual effects.