Neighborhoods,cars, and commuting in New York City: A discrete choice approach

Cities around the world are trying out a multitude of transportation policy and investment alternatives with the aim of reducing car-induced externalities. However, without a solid understanding of how people make their transportation and residential location choices, it is hard to tell which of these policies and investments are really doing the job and which are wasting precious city resources. The focus of this paper is the determinants of car ownership and car use for commuting. Using survey data from 1997 to 1998 collected in New York City, this paper uses discrete choice econometrics to estimate a model of the choices of car ownership and commute mode while also modeling the related choice of residential location.The main story told by this analysis is that New Yorkers are more sensitive to changes in travel time than they are to changes in travel cost. The model predicts that the most effective ways to reduce both auto ownership and car commuting involve changing the relative travel times for cars and transit, making transit trips faster by increasing both the frequency and the speed of service and making auto trips slower – perhaps simply by allowing traffic congestion. Population density also appears to have a substantial effect on car ownership in New York.     

Misclassification in travel surveys and implications to choice modeling: application to household auto ownership decisions

Travel surveys that elicit responses to questions regarding daily activity and travel choices form the basis for most of the transportation planning and policy analysis. The response variables collected in these surveys are prone to errors leading to mismeasurement or misclassification. Standard modeling methods that ignore these errors while modeling travel choices can lead to biased parameter estimates. In this study, methods available in the econometrics literature were used to quantify and assess the impact of misclassification errors in auto ownership choice data. The results uncovered significant misclassification rates ranging from 1 to 40% for different auto ownership alternatives. Also, the results from latent class models provide evidence for variation in misclassification probabilities across different population segments. Models that ignore misclassification were not only found to have lower statistical fit but also significantly different elasticity effects for choice alternatives with high misclassification probabilities. The methods developed in this study can be extended to analyze misclassification in several response variables (e.g., mode choice, activity purpose, trip/tour frequency, and mileage) that constitute the core of advanced travel demand models including tour and activity-based models.

     

Modeling residential sorting effects to understand the impact of the built environment on commute mode choice

This paper presents an examination of the significance of residential sorting or self selection effects in understanding the impacts of the built environment on travel choices. Land use and transportation system attributes are often treated as exogenous variables in models of travel behavior. Such models ignore the potential self selection processes that may be at play wherein households and individuals choose to locate in areas or built environments that are consistent with their lifestyle and transportation preferences, attitudes, and values. In this paper, a simultaneous model of residential location choice and commute mode choice that accounts for both observed and unobserved taste variations that may contribute to residential self selection is estimated on a survey sample extracted from the 2000 San Francisco Bay Area household travel survey. Model results show that both observed and unobserved residential self selection effects do exist; however, even after accounting for these effects, it is found that built environment attributes can indeed significantly impact commute mode choice behavior. The paper concludes with a discussion of the implications of the model findings for policy planning.

A comprehensive dwelling unit choice model accommodating psychological constructs within a search strategy for consideration set formation

This study adopts a dwelling unit level of analysis and considers a probabilistic choice set generation approach for residential choice modeling. In doing so, we accommodate the fact that housing choices involve both characteristics of the dwelling unit and its location, while also mimicking the search process that underlies housing decisions. In particular, we model a complete range of dwelling unit choices that include tenure type (rent or own), housing type (single family detached, single family attached, or apartment complex), number of bedrooms, number of bathrooms, number of storeys (one or multiple), square footage of the house, lot size, housing costs, density of residential neighborhood, and commute distance. Bhat’s (2015) generalized heterogeneous data model (GHDM) system is used to accommodate the different types of dependent outcomes associated with housing choices, while capturing jointness caused by unobserved factors. The proposed analytic framework is applied to study housing choices using data derived from the 2009 American Housing Survey (AHS), sponsored by the Department of Housing and Urban Development (HUD) and conducted by the U.S. Census Bureau. The results confirm the jointness in housing choices, and indicate the superiority of a choice set formation model relative to a model that assumes the availability of all dwelling unit alternatives in the choice set.     

Understanding adolescents’ intentions to commute by car or bicycle as adults

This study focuses on the intentions of adolescents to commute by car or bicycle as adults. The behavioral model is based on intrapersonal and interpersonal constructs from the theory of planned behavior extended to include constructs from the institutional, community and policy domains. Data from a survey among Danish adolescents is analyzed. It is found that car use intentions are related to positive car passenger experience, general interest in cars, and car ownership norms, and are negatively related to willingness to accept car restrictions and perceived lack of behavioral control. Cycling intentions are related to positive cycling experience, willingness to accept car restrictions, negative attitudes towards cars, and bicycle-oriented future vision, and are negatively related to car ownership norms. Attitudinal constructs are related to individual characteristics, such as gender, residential location, current mode choice to daily activities, and parental travel patterns.     

Joint analysis of the spatial impacts of built environment on car ownership and travel mode choice

Concerns over transportation energy consumption and emissions have prompted more studies into the impacts of built environment on driving-related behavior, especially on car ownership and travel mode choice. This study contributes to examine the impacts of the built environment on commuter’s driving behavior at both spatial zone and individual levels. The aim of this study is threefold. First, a multilevel integrated multinomial logit (MNL) and structural equation model (SEM) approach was employed to jointly explore the impacts of the built environment on car ownership and travel mode choice. Second, the spatial context in which individuals make the travel decisions was accommodated, and spatial heterogeneities of car ownership and travel mode choice across traffic analysis zones (TAZs) were recognized. Third, the indirect effects of the built environment on travel mode choice through the mediating variable car ownership were calculated, in other words, the intermediary nature of car ownership was considered. Using the Washington metropolitan area as the study case, the built environment measures were calculated for each TAZ, and the commuting trips were drawn from the household travel survey in this area. To estimate the model parameters, the robust maximum likelihood (MLR) method was used. Meanwhile, a comparison among different model structures was conducted. The model results suggest that application of the multilevel integrated MNL and SEM approach obtains significant improvements over other models. This study give transportation planners a better understanding on how the built environment influences car ownership and commuting mode choice, and consequently develop effective and targeted countermeasures.     

Evaluating household residential preferences for walkability and accessibility across three U.S. regions

Accessibility has been established as a major planning goal in recent years. However, little knowledge exists regarding how individuals value walkability, transit accessibility, and auto accessibility differently when deciding where to live. To fill this knowledge gap, this study conducts residential location choice modeling across three U.S. regions—Atlanta, Puget Sound, and Southeast Michigan. I find that, overall, all three types of accessibility are important determinants of residential location choice. Transit accessibility has a statistically significant positive influence on residential location choice across all three regions. On auto accessibility, results show that commute time by auto has the greatest influence on residential location choice among all independent variables, but auto accessibility to nonwork destinations appears to be inconsequential. Moreover, walkability is found to be a key determinant of residential location choice in the Puget Sound region but not the other two regions. I argue that these regional differences result from a lack of choice among Atlanta and Southeast Michigan residents, that is, a undersupply of walkable neighborhoods inhibits households in the two regions from living in such neighborhoods. This finding suggests the need for cities and regions to promote pedestrian-oriented development in order to broaden residential choice. The results further imply that, due to housing-supply constraints, households often have to live in a neighborhood with a level of accessibility lower than what they prefer. Transportation and land-use planners should address this “residential dissonance” when applying residential location choice models to predict land-use growth patterns.     

Hierarchical ordering of nests in a joint mode and destination choice model

This paper seeks to explore the relationship between mode and destination choice in an integrated nested choice model. A fundamental argument can be made that in certain circumstances, the ordering of choices should be reversed from the usual sequence of destination choice preceding mode choice. This results in a travel demand model where travelers are more likely to change destinations than to change transportation modes. For small and medium size urban areas, particularly in the United States, with less well developed public transit systems that draw few choice riders, this assumption makes much more sense than the traditional modeling assumptions. The models used in the new travel modeling system developed for Knoxville, Tennessee utilize this reversed ordering, with generally good results, which required no external tinkering in the logsum parameters.     

Implicit choice set generation in discrete choice models: Application to household auto ownership decisions

Latent choice set models that account for probabilistic consideration of choice alternatives during decision making have long existed. The Manski model that assumes a two-stage representation of decision making has served as the standard workhorse model for discrete choice modeling with latent choice sets. However, estimation of the Manski model is not always feasible because evaluation of the likelihood function in the Manski model requires enumeration of all possible choice sets leading to explosion for moderate and large choice sets. In this study, we propose a new group of implicit choice set generation models that can approximate the Manski model while retaining linear complexity with respect to the choice set size. We examined the performance of the models proposed in this study using synthetic data. The simulation results indicate that the approximations proposed in this study perform considerably well in terms of replicating the Manski model parameters. We subsequently used these implicit choice set models to understand latent choice set considerations in household auto ownership decisions of resident population in the Southern California region. The empirical results confirm our hypothesis that certain segments of households may only consider a subset of auto ownership levels while making decisions regarding the number of cars to own. The results not only underscore the importance of using latent choice models for modeling household auto ownership decisions but also demonstrate the applicability of the approximations proposed in this study to estimate these latent choice set models.     

A joint flexible econometric model system of household residential location and vehicle fleet composition/usage choices

Modeling the interaction between the built environment and travel behavior is of much interest to transportation planning professionals due to the desire to curb vehicular travel demand through modifications to built environment attributes. However, such models need to take into account self-selection effects in residential location choice, wherein households choose to reside in neighborhoods and built environments that are conducive to their lifestyle preferences and attitudes. This phenomenon, well-recognized in the literature, calls for the specification and estimation of joint models of multi-dimensional land use and travel choice processes. However, the estimation of such model systems that explicitly account for the presence of unobserved factors that jointly impact multiple choice dimensions is extremely complex and computationally intensive. This paper presents a joint GEV-based logit regression model of residential location choice, vehicle count by type choice, and vehicle usage (vehicle miles of travel) using a copula-based framework that facilitates the estimation of joint equations systems with error dependence structures within a simple and flexible closed-form analytic framework. The model system is estimated on a sample derived from the 2000 San Francisco Bay Area Household Travel Survey. Estimation results show that there is significant dependency among the choice dimensions and that self-selection effects cannot be ignored when modeling land use-travel behavior interactions.     

Built environmental impacts on commuting mode choice and distance: Evidence from Shanghai

The impacts of the built environment characteristics in residential neighborhoods on commuting behavior are explored in the literature. Scant evidence, however, is provided to scrutinize the role of the built environment characteristics at job locations. Studies also overlooked the potential error correlations between commuting mode and commuting distance due to the unobserved factors that influence both variables. We examined the impacts of the built environment characteristics at both residential and job locations on commuting mode and distance, by applying a discrete-continuous copula-based model on 857 workers in Shanghai. In contrast with studies of Western countries, we showed residential built environment characteristics are more influential on commute behavior than the built environment characteristics at job locations. This suggests the importance of local specificity in policymaking process. We also found the proportion of four-way intersections, road density, and population density in residential areas are negatively associated with driving probability, with elasticity amounts of −1.00, −0.23, and −0.08, respectively. Hence, dense and pedestrian- and cyclist-oriented development help to reduce travel distance and encourage walking, biking, and transit modes of travel.     

Household-level commuting mode choices,car allocation and car ownership level choices of two-worker households: the case of the city of Toronto

The paper presents a comprehensive investigation on household level commuting mode, car allocation and car ownership level choices of two-worker households in the City of Toronto. A joint econometric model and a household travel survey dataset are used for empirical investigations. Empirical models reveal that significant substitution patterns exist between auto driving and all other mode choices in two-worker households. It is revealed that, female commuters do not prefer auto driving, but in case of a one car (and two commuters with driving licenses) household, a female commuter gets more preference for auto driving option than the male commuter. Reverse commuting (commuting in opposite direction of home to central business district) plays a critical role on household level car allocation choices and in defining the stability of commuting behaviour of two-worker households. Two worker households in higher income zones and with longer commuting distances tend to have higher car ownership levels than others. However, higher transit accessibility to jobs reduces household car ownership levels. The study reveals that both increasing two worker households and reverse commuting would increase dependency on private car for commuting.     

Analysis of vehicle ownership evolution in Montreal,Canada using pseudo panel analysis

This paper employs a pseudo-panel approach to study vehicle ownership evolution in Montreal region, Canada using cross-sectional origin–destination survey datasets of 1998, 2003 and 2008. Econometric modeling approaches that simultaneously accommodate the influence of observed and unobserved attributes on the vehicle ownership decision framework are implemented. Specifically, we estimate generalized versions of the ordered response model—including the generalized, scaled- and mixed-generalized ordered logit models. Socio-demographic variables that impact household’s decision to own multiple cars include number of full and part-time working adults, license holders, middle aged adults, retirees, male householders, and presence of children. Increased number of bus stops, longer bus and metro lengths within the household residential location buffer area decrease vehicle fleet size of households. The observed results also varied across years as manifested by the significance of the interaction terms of some of the variables with the time elapsed since 1998 variable. Moreover, variation due to unobserved factors are captured for part-time working adults, number of bus stops, and length of metro lines. In terms of the effect of location of households, we found that some neighborhoods exhibited distinct car ownership temporal dynamics over the years.     

An integrated analysis of workers’ physically active activity and active travel choice behavior

Using Texas add-on sample data from the 2009 National Household Travel Survey, this study examines adult workers’ daily active choice decisions in the context of physical activity and attendant health benefits. The study looked at workers’ two choice behaviors: active activity and active travel. The first choice behavior, active activity, is developed as an ordered-response model based on the number of physically active recreational activities pursued during the workday. The second choice behavior, active travel, is developed as a binary-response model that examines workers’ active travel choices—whether or not the worker used any active mode of travel during the same workday. The study improves the understanding and knowledge of observed factors influencing workers’ physically active activity-travel behavior. The study also provides several observations regarding the role (and constraints) of employment in individuals’ active choices. Using a flexible copula modeling methodology, we explore the true correlation (or dependence) between the two behavior choices that could occur due to the presence of unobserved factors, suggesting a simultaneously low or simultaneously high propensity for being physically active across workers. The study findings suggest that transportation and public health policy makers can mutually benefit from encouraging workers to be physically active (from an activity and/or travel perspective). Overall, the study draws attention to the integrated nature of the public health and transportation fields, thereby providing a distinct view of active/inactive choice behavior. To our knowledge, this is the first study exploring a rich variety of components for workers’ active activity-travel behavior through a robust copula approach.     

Immigration, residential location, car ownership, and commuting behavior: a multivariate latent class analysis from California

Using latent class cluster analysis, this paper investigates the spatial, social, demographic, and economic determinants of immigrants’ joint distribution among travel time, mode choice, and departure time for work using the 2000 Census long form data. Through a latent tree structure analysis, age, residential location, immigration stage, gender, personal income, and race are found to be the primary determinants in the workplace commute decision-making process. By defining several relatively homogeneous population segments, the likelihood of falling into each segment is found to differ across age groups and geography, with different indicators affecting each group differentially. This analysis complements past studies that used regression models to investigate socio-demographic indicators and their impact on travel behavior in two distinct ways: (a) analysis is done by considering travel time, mode choice, and departure time for work simultaneously, and (b) heterogeneity in behavior is accounted for using methods that identify different groups of behavior and then their determinants. Conclusively the method here is richer than many other methods used to study the ethnically diverse population of California and shows the addition of geographic location and latent segment identification to greatly improve our understanding of specific behaviors. It also provides evidence that immigrants are as diverse as the non-immigrant population and transportation policies need to be defined accordingly.

On accommodating spatial interactions in a Generalized Heterogeneous Data Model (GHDM) of mixed types of dependent variables

We develop an econometric framework for incorporating spatial dependence in integrated model systems of latent variables and multidimensional mixed data outcomes. The framework combines Bhat's Generalized Heterogeneous Data Model (GHDM) with a spatial (social) formulation to parsimoniously introduce spatial (social) dependencies through latent constructs. The applicability of the spatial GHDM framework is demonstrated through an empirical analysis of spatial dependencies in a multidimensional mixed data bundle comprising a variety of household choices – household commute distance, residential location (density) choice, vehicle ownership, parents’ commute mode choice, and children's school mode choice – along with other measurement variables for two latent constructs – parent's safety concerns about children walking/biking to school and active lifestyle propensity. The GHDM framework identifies an intricate web of causal relationships and endogeneity among the endogenous variables. Furthermore, the spatial (social) version of the GHDM model reveals a high level of spatial (social) dependency in the latent active lifestyle propensity of different households and moderate level of spatial dependency in parents’ safety concerns. Ignoring spatial (social) dependencies in the empirical model results in inferior data fit, potential bias and statistical insignificance of the parameters corresponding to nominal variables, and underestimation of policy impacts.     

Role of the built environment on mode choice decisions: additional evidence on the impact of density

Density is a key component in the recent surge of mixed-use neighborhood developments. Empirical research has shown an inconsistent picture on the impact of density. In particular, it is unclear whether it is the density or the variables that go long with density that affect people’s travel behavior. Many existing studies on density neglect confounding factors, for example, residential self-selection, generalized travel cost, accessibility, and access to transit stations. In addition, most still use a single trip as their observation unit, even though trip chaining is well recognized. The goal of this paper is to assess the role of density in affecting mode choice decisions in home-based work tours, while controlling for confounding factors. Using the dataset collected in the New York Metropolitan Region, we estimated a simultaneous two-equation system comprising two mutually interacting dependent variables: car ownership and the propensity to use auto. The results confirm the role of density after controlling for the confounding factors; in particular, employment density at work exerts more influence than residential density at home. The study also demonstrates the importance of using tour as the analysis unit in mode choice decisions. The study advances the field by analyzing the role of the built environment on home-based work tours. New knowledge is obtained in the relative contribution of density vs. a set of correlated factors, including generalized travel cost, accessibility, and access to transit stations.

Using demand elasticities from disaggregate mode choice models

Three of the most highly regarded disaggregate mode split models incorporate very different estimates of the responsiveness, or elasticity, of mode choice to changes in auto travel times and costs. These differences appear to be due in part to the varying specifications used by the model, and particularly whether certain variables (such as a dummy variable for CBD destinations or automobile ownership) are included in addition to the more traditional variables (such as travel time, cost, and household income). More research is needed on the implications of the theory of traveler choices for model specification and the effect of alternative, but theoretically justifiable, specifications on elasticity estimates. Until this research reduces our uncertainty about the elasticity of demand, analysts evaluating transportation policies should assess the sensitivity of their results to the range of plausible elasticities or models.     

A note on consistency of choice units in entropy models

In using entropy maximization models to forecast locational and travel behaviour, one is confronted with the problem of delineating the choice process as precisely as possible. In addition to defining a fine-grain choice structure implying individuals seeking distinct location sites within residential zones and travelling to distinct jobs or shops within destination zones, this note also accounts for the fact that the location choice is of a site for a household or firm, but the corresponding travel choices are by individual members of a household. In conjunction with disaggregation across quantities with large variance, the above principles are applied to formulate improved versions of residential and shopping location models.