Spatial heterogeneous impact of built environment on household auto ownership levels: evidence from analysis at traffic analysis zone scales

Abstract

Household auto ownership modeling plays an important role in travel demand analysis because it reveals a key link between travel behavior and built environment. The object of this paper is to estimate the impact of built environment on household auto ownership levels by using a multilevel, mixed ordered probit model, which captures the spatial heterogeneity across traffic analysis zones (TAZs). Parameters are estimated with Bayesian estimation method via Markov Chain Monte Carlo (MCMC) sampling. The data come from the household travel survey 2007–2008 for the Washington–Baltimore Region. Empirical results indicate that our multilevel ordered probit model can effectively address the spatial heterogeneity across zones and provide better model fit than traditional ordered probit model. It is found that built environment can explain about 42.8% of the spatial heterogeneity in household auto ownership. Our findings can help policy makers develop a better understanding of how the built environment can influence household auto ownership.

Keywords:

• Built environment
• Auto ownership
• Ordered probit model
• Multilevel analysis
• Spatial heterogeneity
• Bayesian method

Acknowledgements

This work is supported by the National Basic Research Program of China (2012CB725404).