Online Program Home
My Program

Abstract Details

Activity Number: 604 - Bayesian Inference in Discrete Choice Analysis of Consumer Behavior
Type: Topic Contributed
Date/Time: Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
Sponsor: Business and Economic Statistics Section
Abstract #306733 Presentation
Title: A Model for Built Environment Effects on Mode Usages
Author(s): Kai Yoshioka* and Tomomi Miyazaki
Companies: University of California, Irvine and Kobe University
Keywords: Discrete Choice; Potential Outcomes; Transportation Economics; Simulation-based Inference; Bayesian Econometrics; MCMC

This paper develops an econometric framework for estimating the effect of the built environment on transportation mode usages when a large fraction of the population under study is nonlicensed. We use a multivariate ordinal model with a binary selection component to (i) allow for heterogeneous built environment effects and heterogeneous unobserved substitution patterns across the nonlicensed and licensed groups, and (ii) account for the indirect effect urban form has on mode usages via driver's license choice. This exposition focuses on the joint modeling of correlated and discrete outcomes (binary and ordinal), strategizing with identification restrictions and nonidentification, and the efficient estimation of model parameters. We use our econometric framework to study the effect of the built environment on the travel habits of the Japanese elderly. Built environment effects are found to be similar across groups with a few stark exceptions. Consistent with studies based on the United States, elasticities are found to be nonzero but modest at best.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program