Subodh Dubey

Development of Vehicle Fleet Composition Model System for Implementation in Activity-Based Travel Model

The development of a vehicle fleet composition and utilization model system that may be incorporated into a larger activity-based travel demand model is described. It is of interest and importance to model household vehicle fleet composition and utilization behavior because the energy and environmental impacts of personal travel are dependent not only on the number of vehicles but also on the mix of vehicles that a household owns and the extent to which different vehicles are used. A vehicle composition (fleet mix) and utilization model system was developed for integration into the activity-based travel demand model that was being developed for the greater Phoenix metropolitan area in Arizona. At the heart of the vehicle fleet mix model system is a multiple discrete continuous extreme value model capable of simulating vehicle ownership and use patterns of households. Vehicle choices are defined by a combination of vehicle body type and age category and the model system is capable of predicting vehicle composition and utilization patterns at the household level. The model system is described and results are presented of a validation and policy sensitivity analysis exercise demonstrating the efficacy of the model.

A New Spatial Multiple Discrete‐Continuous Modeling Approach to Land Use Change Analysis

Land-use change models are used in a variety of fields such as planning, urban science, ecological science, climate science, geography, watershed hydrology, environmental science, political science, and transportation to examine future land-use scenarios as well as to evaluate the potential effects of policies directed toward engendering a socially or economically or ecologically desirable pattern of future land-use that minimizes negative externalities. More recently, there has been substantial attention in the scientific literature on biodiversity loss, deforestation consequences, and carbon emissions increases caused by patterns of urban and rural land-use development, and associated climate change impacts . In this paper, we contribute to the vibrant and interdisciplinary literature on land-use analysis by proposing a new econometric approach to specify and estimate a model of land-use change that is capable of predicting both the type and intensity of urban development patterns over large geographic areas, while also explicitly acknowledging geographic proximity-based spatial dependencies in these patterns. As such, the motivations of this paper stem both from an empirical perspective as well as a methodological perspective. At an empirical level , the paper models land-use in multiple discrete states, along with the area invested in each land-use discrete state, within each spatial unit in an entire urban region. The spatial unit of analysis is a quarter-of-a-mile square grid, within which there can be multiple land-uses with associated land areas. At a methodological level , the paper focuses on specifying and estimating a spatial multiple discrete-continuous (MDC) probit model. To our knowledge, this is the first formulation and attempt to include spatial dependency patterns originating from both the systematic component (sometimes referred to as “spillover effects”) as well as spatial effects originating from the unobserved component (in the literature, it is typical to use the label “spatial” only if the latter effects are accommodated) in MDC models. . Further, the two dominant techniques, both based on simulation methods, for the estimation of standard discrete choice models with spatial dependence are the frequentist recursive importance sampling (RIS) estimator and the Bayesian Markov Chain Monte Carlo (MCMC)-based estimator. However, both of these methods are confronted with multi-dimensional normal integration, and are cumbersome to implement in typical empirical contexts with even moderate estimation sample sizes. In the current paper, we show how Bhat’s maximum approximate composite marginal likelihood (MACML) inference approach can be gainfully applied for the estimation of a spatial multiple discrete-continuous probit (MDCP) model. This method is easy to implement, require no simulation, and involve only univariate and bivariate cumulative normal distribution function evaluations, regardless of the number of alternatives or the number of choice occasions per observation unit, or the number of observation units, or the nature of social/spatial dependence structures. The spatial MDCP formulation also accommodates spatial heterogeneity and heteroscedasticity in the dependent variable, and should be applicable in a wide variety of fields where social and spatial dependencies between decision agents (or observation units) lead to spillover effects in multiple discrete-continuous choices (or states).