Inference on mode preferences,vehicle purchases,and the energy paradox using a Bayesian structural choice model

Discrete choice modeling is experiencing a reemergence of research interest in the inclusion of latent variables as explanatory variables of consumer behavior. There are several reasons that motivate the integration of latent attributes, including better-informed modeling of random consumer heterogeneity and treatment of endogeneity. However, current work still is at an early stage and multiple simplifying assumptions are usually imposed. For instance, most previous applications assume all of the following: independence of taste shocks and of latent attributes, exclusion restrictions, linearity of the effect of the latent attributes on the utility function, continuous manifest variables, and an a priori bound for the number of latent constructs. We derive and apply a structural choice model with a multinomial probit kernel and discrete effect indicators to analyze continuous latent segments of travel behavior, including inference on the energy paradox. Our estimator allows for interaction and simultaneity among the latent attributes, residual correlation, nonlinear effects on the utility function, flexible substitution patterns, and temporal correlation within responses of the same individual. Statistical properties of the Bayes estimator that we propose are exact and are not affected by the number of latent attributes.     

Is car drivers’ response to congestion charging schemes based on the correct perception of price signals?

This paper deals with the question of whether the capability of car drivers to estimate the cost of a new hypothetical, highly differentiated congestion charge influences their decision to change travel behaviour. The analysis makes use of an integrated choice and latent variable model (ICLV) which merges classic choice models with the structural equation approach (SEM) for latent variables. This hybrid model improves the explanatory power considerably compared with a conventional discrete choice model. The results suggest that charge complexity decreases the resistance in considering behavioural changes. Car drivers tend to avoid a travel option where the price is not known beforehand, a phenomenon known as ambiguity avoidance.     

The relation between bicycle commuting and non-work cycling: results from a mobility panel

This study aims to establish whether or not bicycle commuting and cycling for other purposes (e.g. shopping, visiting friends) are related over time. Using previously gathered panel data (the Dutch mobility panel) these relationships are revealed by (1) a series of conditional change models and (2) a latent transition model. The conditional change models indicate that, with a lag of 1 year and controlling for a range of background characteristics, bicycle commuting and non-work cycling (in number of weekly trips) have a positive reciprocal influence on each other. The models show that work-related factors, such as the distance to work or whether a person receives a travel allowance, affect not only bicycle commuting but also non-work cycling. The latent transition model indicates that people can be clustered into four groups: non-cyclists, non-work cyclists, all-around cyclists and commuter cyclists. This model shows that people with a consistent propensity to not cycle at all (non-cyclists) or to cycle for both work and non-work purposes (all-around cyclists) are most stable in their travel behavior. Non-work cyclists and commuter cyclists are less stable in travel behavior. The model also shows that all-around cyclists are not (significantly) affected by a change in the distance to work. The article concludes with several directions for future research.     

Latent variables and route choice behavior

In the last decade, a broad array of disciplines has shown a general interest in enhancing discrete choice models by considering the incorporation of psychological factors affecting decision making. This paper provides insight into the comprehension of the determinants of route choice behavior by proposing and estimating a hybrid model that integrates latent variable and route choice models. Data contain information about latent variable indicators and chosen routes of travelers driving regularly from home to work in an urban network. Choice sets include alternative routes generated with a branch and bound algorithm. A hybrid model consists of measurement equations, which relate latent variables to measurement indicators and utilities to choice indicators, and structural equations, which link travelers’ observable characteristics to latent variables and explanatory variables to utilities. Estimation results illustrate that considering latent variables (i.e., memory, habit, familiarity, spatial ability, time saving skills) alongside traditional variables (e.g., travel time, distance, congestion level) enriches the comprehension of route choice behavior.     

Estimating heterogeneity of car travelers on mode shifting behavior based on discrete choice models

In order to understand the mode shift behavior of car travelers and relieve traffic congestion, a Stated Preference survey has been conducted in the city of Ji'nan in China to analyze bus choice behavior and the heterogeneity of car travelers. Several discrete choice models, including multinomial logit, mixed logit and latent class model (LCM) are developed based on these survey data. A comparative analysis indicates that the LCM has the highest precision and is more suitable to analyze the heterogeneity of car travelers. The LCM divides car travelers into three classes. Different classes have different sets of influencing factors in the model. Policy recommendations are also proposed for those classes to promote bus shift from car travelers based on the model results. Finally, sensitivity analysis on parking fees and fuel cost is carried out on the LCMs under different bus service levels. Car travelers have different sensitivities to the influencing factors. The conclusions indicate that the LCM can reflect the heterogeneity and preferences of car travelers and can be used to understand how to shift the behavior of car travelers and make more effective traffic policy.     

Synopsis of bicycle demand in the City of Toronto: Investigating the effects of perception,consciousness and comfortability on the purpose of biking and bike ownership

This paper provides an empirical basis for the evaluation of policies and programs that can increase the usage of bikes for different purposes as well as bike ownership. It uses an integrated econometric model of latent variable connecting multiple discrete choices. Empirical models are estimated by using a bicycle demand survey conducted in the City of Toronto in 2009. Empirical investigations reveal that latent perceptions of ‘bikeability’ and ‘safety consciousness’ directly influence the choice of biking. It is also found that the choice of the level of bike ownership (number of bikes) is directly influenced by latent ‘comfortability of biking’. The number of bikes owned moreover has a strong influence on the choices of biking for different purposes. It is clear that bike users in the City of Toronto are highly safety conscious. Increasing on-street and separate bike lanes proved to have the maximum effects on attracting more people to biking by increasing the perception of bikeability in the city, comfortability of biking in the city and increasing bike users’ sense of safety. In terms of individuals’ characteristics, older males are found to be the most conformable and younger females are the least comfortable group of cyclists in Toronto.     

A bi-objective cyclist route choice model

It is widely acknowledged that cyclists choose their route differently to drivers of private vehicles. The route choice decision of commuter drivers is often modelled with one objective, to reduce their generalised travel cost, which is a monetary value representing the combined travel time and vehicle operating cost. Commuter cyclists, on the other hand, usually have multiple incommensurable objectives when choosing their route: the travel time and the suitability of a route. By suitability we mean non-subjective factors that characterise the suitability of a route for cycling, including safety, traffic volumes, traffic speeds, presence of bicycle lanes, whether the terrain is flat or hilly, etc. While these incommensurable objectives are difficult to be combined into a single objective, it is also important to take into account that each individual cyclist may prioritise differently between travel time and suitability when they choose a route.This paper proposes a novel model to determine the route choice set of commuter cyclists by formulating a bi-objective routing problem. The two objectives considered are travel time and suitability of a route for cycling. Rather than determining a single route for a cyclist, we determine a choice set of optimal alternative routes (efficient routes) from which a cyclist may select one according to their personal preference depending on their perception of travel time versus other route choice criteria considered in the suitability index. This method is then implemented in a case study in Auckland, New Zealand.The study provides a starting point for the trip assignment of cyclists, and with further research, the bi-objective routing model developed can be applied to create a complete travel demand forecast model for cycle trips. We also suggest the application of the developed methodology as an algorithm in an interactive route finder to suggest efficient route choices at different levels of suitability to cyclists and potential cyclists.     

Who cycles more? Determining cycling frequency through a segmentation approach in Montreal,Canada

The decision to cycle frequently in an urban setting is a complex process and is affected by a variety of factors. This study analyzed the various factors influencing cycling frequency among 1707 cyclists from Montreal, Canada using an ordinal logistic regression. A segmentation of cyclists is used in a series of ordinal logistic models to better understand the different impacts of variables on the frequency of cycling among each group of cyclists for commute and for utilitarian purposes. Our models show a variation in the impacts of each dependent variable on frequency of cycling across the various segments of cyclists. Mainly making cyclists feel safe not only on bicycle specific infrastructure but also on regular streets, emphasizing the low cost, convenience and improving the opinion on cycling in the population are effective interventions to increase bicycle usage. Also, it was shown that women were less likely to cycle to work than men, but more likely to cycle for other utilitarian trips, pointing at the presence of specific barriers to commuting for woman. Although the findings from this study are specific to Montreal, they can be of interest to transportation planners and engineers working toward increasing cycling frequency in other regions.     

Investigating the subjective and objective factors influencing teenagers’ school travel mode choice – An integrated choice and latent variable model

In this paper, we apply Bhat and Dubey’s (2014) new probit-kernel based Integrated Choice and Latent Variable (ICLV) model formulation to analyze children’s travel mode choice to school. The new approach offered significant advantages, as it allowed us to incorporate three latent variables with a large data sample and with 10 ordinal indicators of the latent variables, and still estimate the model without any convergence problems. The data used in the empirical analysis originates from a survey undertaken in Cyprus in 2012. The results underscore the importance of incorporating subjective attitudinal variables in school mode choice modeling. The results also emphasize the need to improve bus and walking safety, and communicate such improvements to the public, especially to girls and women and high income households. The model application also provides important information regarding the value of investing in bicycling and walking infrastructure.     

Development of an indicator to assess the spatial fit of discrete choice models

Discrete choice models are increasingly implemented using geographical data. When this is the case, it may not be sufficient to project market shares accurately, but also to correctly replicate the spatial pattern of choices. Analysts might then be interested in assessing the results of a model’s fit relative to the spatial distribution of the observed responses. While canonical approaches exist for the exploratory spatial analysis of continuous variables, similar tools have not been widely implemented for discrete choice models, where the variable of interest is categorical. For this reason, despite recent progress with spatial models for discrete outcomes, there is still not a simple and intuitive tool to assess the quality of the spatial fit of a discrete choice model. The objective of this paper is to introduce a new indicator of spatial fit that can be applied to the results of discrete choice models. Utility of the indicator is explored by means of numerical experiments and then demonstrated by means of a case study of vehicle ownership in Montreal, Canada.     

How does air pollution influence cycling behaviour? Evidence from Beijing

It is widely believed air pollution is an obstacle to cycling as it has negative effects on cyclists’ health outcomes and deteriorates their cycling experiences. However, the empirical studies investigating the impact of air pollution on cycling behaviour remains scarce. The aim of this paper is to fill the gap by looking at Beijing as a case study. The authors conducted a survey of 307 cyclists on the days with different levels of air quality in terms of concentration of PM_2.5 in 2015. The results show that in the polluted weather, those who persist in cycling are more likely to be male, over 30 years old, lower income or those who travel short distances. Specifically, female cyclists have a higher tendency to shift from cycling to public transit than the males and medium and high-income earners are more likely to shift to using a car than low income earners. The residents’ subjective perceptions of safety and comfort have major effects on their cycling behaviour. A higher perception of comfort and safety is related to a higher possibility of continuing cycling when air quality became polluted. Cycling for commuting trips is less likely to be replaced by other modes than cycling for non-commuting trips, such as shopping. Results of this study reveal that improving air quality in a metropolitan area such as Beijing has co-benefits of cycling renaissance. The huge investments into cycling infrastructure should be integrated with policies designed to create an attractive environment for cycling.     

The evaluation of high occupancy vehicle lanes on sun yat‐sen freeway in Taiwan

This study proposes a methodological framework to incorporate latent factors, including direct and indirect perceptions, as the explanatory variables in a discrete choice models using revealed preference and stated preference data sets. The methodology requires the estimation of a model system comprising of a discrete choice model and the structural and measurement equations of a latent variable model. The application involves the evaluation of responses to the new high occupancy vehicle (HOV) lanes on the Sun Yat‐Sen Freeway in Taiwan. The results obtained from this study provide valuable insights into the planning and assessment of HOV lanes.     

Accounting for stochastic variables in discrete choice models

The estimation of discrete choice models requires measuring the attributes describing the alternatives within each individual’s choice set. Even though some attributes are intrinsically stochastic (e.g. travel times) or are subject to non-negligible measurement errors (e.g. waiting times), they are usually assumed fixed and deterministic. Indeed, even an accurate measurement can be biased as it might differ from the original (experienced) value perceived by the individual.Experimental evidence suggests that discrepancies between the values measured by the modeller and experienced by the individuals can lead to incorrect parameter estimates. On the other hand, there is an important trade-off between data quality and collection costs. This paper explores the inclusion of stochastic variables in discrete choice models through an econometric analysis that allows identifying the most suitable specifications. Various model specifications were experimentally tested using synthetic data; comparisons included tests for unbiased parameter estimation and computation of marginal rates of substitution. Model specifications were also tested using a real case databank featuring two travel time measurements, associated with different levels of accuracy.Results show that in most cases an error components model can effectively deal with stochastic variables. A random coefficients model can only effectively deal with stochastic variables when their randomness is directly proportional to the value of the attribute. Another interesting result is the presence of confounding effects that are very difficult, if not impossible, to isolate when more flexible models are used to capture stochastic variations. Due the presence of confounding effects when estimating flexible models, the estimated parameters should be carefully analysed to avoid misinterpretations. Also, as in previous misspecification tests reported in the literature, the Multinomial Logit model proves to be quite robust for estimating marginal rates of substitution, especially when models are estimated with large samples.     

Exploring temporal fluctuations of daily cycling demand on Dutch cycle paths: the influence of weather on cycling

In the pursuit of sustainable mobility policy makers are giving more attention to cycling. The potential of cycling is shown in countries like the Netherlands, where cycling covers 25 % of all person trips. However, the effect of policy interventions on cycling demand is difficult to measure, not least caused by difficulties to control for changing context variables like weather conditions. According to several authors weather has a strong influence on cycling demand, but quantitative studies about the relationship are scarce. We therefore further explored this relationship, with the aim of contributing to the development of a generic demand model with which trend and coincidence in bicycle flows might be unraveled. The study is based on time-series between 1987 and 2003 of daily bicycle flows, collected on 16 cycle paths near two cities in the Netherlands. The regression analyses show that, not surprisingly, recreational demand is much more sensitive to weather than utilitarian demand. Most daily fluctuations (80 %) are described by weather conditions, and no less than 70 % of the remaining variation is locally constrained. The regression can therefore mainly be improved by incorporating path specific, as yet unknown, variables. We used the regression results to calculate weather-inclusive bicycle flow predictions and found indications of a downward trend in recreational demand. This trend has been off-set in the observed flows by more favorable weather conditions over the years considered.     

Exploring the role of individual attitudes and perceptions in predicting the demand for cycling: a hybrid choice modelling approach

Cycling is often promoted as a means of reducing urban congestion and improving health, social and environmental outcomes. However, the quantification of these potential benefits is not well established. This is due in part to practical difficulties in estimating cycling demand and a lack of sound methodologies to appraise cycling initiatives. In this paper we attempt to address this need by developing predictive models of cycle demand, relative to other transport modes, that capture not only the impacts of observed characteristics such as age and travel time but also the role of attitudes and perceptions. Using data from a stated preference survey, we estimate a hybrid choice model for cycle use that incorporates the role of attitudes towards cycling, perceptions of the image associated with cycling, and the stress arising from safety concerns. Model results indicate that the latent attitudes and perceptions explain an important part of the non-observable utility in a simple multinomial logit choice model. We also demonstrate policy analysis using the hybrid choice model, which allows comparisons of ‘hard’ policies such as the provision of parking facilities against ‘soft’ measures such as cycle promotion schemes.     

Making use of respondent reported processing information to understand attribute importance: a latent variable scaling approach

In recent years we have seen an explosion of research seeking to understand the role that rules and heuristics might play in improving the predictive capability of discrete choice models, as well as delivering willingness to pay estimates for specific attributes that may (and often do) differ significantly from estimates based on a model specification that assumes all attributes are relevant. This paper adds to that literature in one important way—it explicitly recognises the endogeneity issues raised by typical attribute non-attendance treatments and conditions attribute parameters on underlying unobserved attribute importance ratings. We develop a hybrid model system involving attribute processing and outcome choice models in which latent variables are introduced as explanatory variables in both parts of the model, explaining the answers to attribute processing questions and explaining heterogeneity in marginal sensitivities in the choice model. The resulting empirical model explains how lower latent attribute importance leads to a higher probability of indicating that an attribute was ignored or that it was ranked as less important, as well as increasing the probability of a reduced value for the associated marginal utility coefficient in the choice model. The model does so by treating the answers to information processing questions as dependent rather than explanatory variables, hence avoiding potential risk of endogeneity bias and measurement error.     

Influences on bicycle use

A stated preference experiment was performed in Edmonton in Canada to both examine the nature of various influences on bicycle use and obtain ratios among parameter values to be used in the development of a larger simulation of household travel behaviour. A total of 1128 questionnaires were completed and returned by current cyclists. Each questionnaire presented a pair of possible bicycle use alternatives and asked which was preferred for travel to a hypothetical all-day meeting or gathering (business or social). Alternatives were described by specifying the amounts of time spent on three different types of cycling facility and whether or not showers and/or secure bicycle parking were available at the destination. Indications of socio-economic character and levels of experience and comfort regarding cycling were also collected. The observations thus obtained were used to estimate the parameter values for a range of different utility functions in logit models representing this choice behaviour. The results indicate, among other things, that time spent cycling in mixed traffic is more onerous than time spent cycling on bike lanes or bike paths; that secure parking is more important than showers at the destination; and that cycling times on roadways tend to become less onerous as level of experience increases. Some of these results are novel and others are consistent with findings regarding bicycle use in work done by others, which is seen to add credence to this work. A review of previous findings concerning influences on cycling behaviour is also included.     

Cycling near misses: Their frequency,impact, and prevention

This paper explores cyclists’ experiences of non-injury incidents, arguing that these are important for cycling experience and uptake as well as for injury prevention. It discusses different types of non-injury incident collected in a recent survey of UK cyclists. These are everyday occurrences that in some cases have a substantially negative impact on cycling experiences. This article explores the impact of different incident types on people cycling both immediately and in the future. It analyses what near misses tell us about cyclists’ experience of problems related to road user behaviour and culture, and infrastructural conditions for cycling. The paper explores what cyclists experiencing near misses think might have prevented them. Based on this and on a comparison with common types of injury incidents, summary recommendations are made for policy and future research.     

Motivators and deterrents of bicycling: comparing influences on decisions to ride

In a survey of 1,402 current and potential cyclists in Metro Vancouver, 73 motivators and deterrents of cycling were evaluated. The top motivators, consistent among regular, frequent, occasional and potential cyclists, were: routes away from traffic noise and pollution; routes with beautiful scenery; and paths separated from traffic. In factor analysis, the 73 survey items were grouped into 15 factors. The following factors had the most influence on likelihood of cycling: safety; ease of cycling; weather conditions; route conditions; and interactions with motor vehicles. These results indicate the importance of the location and design of bicycle routes to promote cycling.     

The impact of weather conditions on bikeshare trips in Washington,DC

Bicycle usage can be affected by colder weather, precipitation, and excessive heat. The research presented here analyzes the effect of weather on the use of the Washington, DC, bikeshare system, exploiting a dataset of all trips made on the system. Hourly weather data, including temperature, rainfall, snow, wind, fog, and humidity levels are linked to hourly usage data. Statistical models linking both number of users and duration of use are estimated. Further, we evaluate trips from bikeshare stations within one quarter mile of Metro (subway) stations at times when Metro is operating. This allows us to determine whether Metro serves as a back-up option when weather conditions are unfavorable for bicycling. Results show that cold temperatures, rain, and high humidity levels reduce both the likelihood of using bikeshare and the duration of trips. Trips taken from bikeshare stations proximate to Metro stations are affected more by rain than trips not proximate to Metro stations and less likely when it is dark. This information is useful for understanding bicycling behavior and also for those planning bikeshare systems in other cities.