ABSTRACT
Disentangling individual contributions to travel distance continues to be an active field of research, with advances in econometrics being germane. In this study, we focus on several econometric techniques, some of which have been applied in isolation in past studies, but the combination of which has yet to be applied to the problem. We apply a series of statistical tests using the method of artificial regression to test the joint effects of functional form and spatial autocorrelation on model fit. An empirical application is made to the Greater Toronto Area using a large-scale travel survey. Several results found in past studies are refuted through the use of advanced econometric methods, including the previous finding of a positive correlation between travel distance and density in the study region. Model results are validated using bootstrapped sampling and local indicators of spatial association.
Acknowledgments
We acknowledge the support of a Natural Sciences and Engineering Research Council of Canada (NSERC) CGS-D Scholarship by the first author and NSERC Discovery Grant by the second author. The authors are grateful for the comments of the reviewers on an earlier version of this article.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Exogenous variables are uncorrelated with the residual random error term. We use the term explanatory to indicate that we are focused on exploring the sign and relative magnitude of effects from many variables rather than causal inference on a single variable.
2 Chowdhury and Scott (Citation2020) find 1 km to be reasonable distance in a similar study of travel distance in Halifax, Canada.