Forecasting Adoption of Ultra-Low-Emission Vehicles Using Bayes Estimates of a Multinomial Probit Model and the GHK Simulator

Daziano, Ricardo A.; Achtnicht, Martin · 2013 · OpenAlex-citations

DOI: 10.1287/trsc.2013.0464

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of forecasting consumer adoption of ultra-low-emission vehicles, such as battery electric and hydrogen-fueled cars, in the European Union. While these technologies offer significant efficiency and emission benefits, their market penetration is hindered by high purchase prices, limited driving range, and insufficient refueling infrastructure. The authors aim to model consumer trade-offs regarding these barriers and provide robust forecasts that account for estimation uncertainty, a critical factor when evaluating emerging technologies. To achieve this, the study employs a multinomial probit model with a fully flexible covariance structure, allowing for complex substitution patterns among different fuel types. The analysis uses stated-preference data from a Germany-wide survey of approximately 600 potential light-duty vehicle buyers, conducted via computer-assisted personal interviewing. Respondents evaluated hypothetical vehicle choices characterized by attributes including purchase price, fuel costs, engine power, CO2 emissions, fuel type, and fuel availability. Methodologically, the authors combine Bayesian estimation using a Gibbs sampler with the GHK (Geweke-Hajivassiliou-Keane) simulator. This approach allows for the computation of the posterior distribution of choice probabilities, enabling the derivation of Bayesian credible intervals for market share forecasts, which addresses the computational infeasibility of frequentist maximum simulated likelihood estimators for this scale of problem. The results demonstrate that the Bayesian point estimates accurately reproduce observed market shares. The study then simulates consumer response to qualitative changes in vehicle attributes, specifically focusing on the expansion of service station networks. The findings indicate that fuel availability is a significant determinant of choice. Notably, if the availability of charging stations for electric vehicles were increased to match the density of conventional fuel stations (100% coverage), the market penetration of electric vehicles would increase by more than threefold. The model also captures the negative impact of higher purchase prices and CO2 emissions on utility, while confirming the positive utility associated with engine power and fuel availability. The significance of this work lies in its methodological contribution to discrete choice modeling and its policy implications. By integrating Bayesian estimation with the GHK simulator, the authors provide a feasible framework for generating forecasts with explicit measures of uncertainty (credible intervals) for complex multinomial probit models. This is particularly valuable for policymakers and manufacturers assessing the impact of infrastructure investments on the adoption of sustainable transportation technologies. The study highlights that expanding charging infrastructure is a critical lever for accelerating the transition to electric vehicles, potentially overcoming the primary barrier of perceived reliability and convenience.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• acceptance adoption