Does the choice model method and/or the data matter?

The opportunity to have seven data sets associated with a stated choice experiment that are very similar in content and design is rare, and provides an opportunity to look in detail at the empirical evidence within and between each data set in the context of a range of discrete choice estimation methods, from multinomial logit to latent class to scale multinomial logit to mixed logit, and the most general model, generalized mixed multinomial logit that accounts for preference and scale heterogeneity. Given the problems associated with data from different countries and time periods, we estimate separate models for each data set, obtaining values of travel time savings that are then updated post estimation to a common dollar for comparative purposes. We also pooled all data sets for a scaled MNL model, treating each data set as a set of three separate utility expressions, but linked to the other data sets through scale heterogeneity. This is not behaviourally appropriate with MNL, latent class or mixed logit. The main question investigated is whether there exists greater synergy in the willingness to pay evidence within model form across data sets compared to across model forms within data sets. The evidence suggests that there is a relatively greater convergence of evidence across the choice models, with the exception of generalized mixed logit, after controlling for data set differences; and there is strong evidence to suggest that differences between data sets do matter.     

Does scale heterogeneity across individuals matter? An empirical assessment of alternative logit models

There is growing interest in establishing a mechanism to account for scale heterogeneity across individuals (essentially the variance of a variance term or the standard deviation of utility over different choice situations), in addition to the more commonly identified taste heterogeneity in mixed logit models. A number of authors have recently proposed a model that recognizes the relationship between scale and taste heterogeneity, and investigated the behavioural implications of accounting for scale heterogeneity in contrast to a term in the utility function, itself. In this paper we present a general model that extends the mixed logit model to explicitly account for scale heterogeneity in the presence of preference heterogeneity, and compare it with models that assume only scale heterogeneity (referred to as the scale heterogeneous multinomial logit model) and only preference heterogeneity. Our empirical assessment suggests that accommodating scale heterogeneity in the absence of accounting for preference heterogeneity may be of limited empirical interest, resulting in a statistically inferior model, despite it being an improvement over the standard MNL model. Scale heterogeneity in the presence of preference heterogeneity does garner favour, with the generalized mixed logit model an improvement over the standard mixed logit model. The evidence herein suggests, however, that compared to a failure to account for preference heterogeneity that is consequential, failure to account for scale heterogeneity may not be of such great empirical consequence in respect of behavioural outputs such as direct elasticities and willingness to pay. However additional studies are required to establish the extent to which this evidence is transferable to a body of studies.     

Accounting for scale heterogeneity within and between pooled data sources

There is growing interest in incorporating both preference heterogeneity and scale heterogeneity in choice models, as a way of capturing an increasing number of sources of utility amongst a set of alternatives. The extension of mixed logit to incorporate scale heterogeneity in a generalised mixed logit (GMXL) model provides a way to accommodate these sources of influence, observed and unobserved. The small but growing number of applications of the GMXL model have parameterized scale heterogeneity as a single estimate; however it is often the case that analysts pool data from more than one source, be it revealed preference (RP) and stated preference (SP) sources, or multiple SP sources, inducing the potential for differences in the scale factor between the data sources. Existing practice has developed ways of accommodating scale differences between data sources by adopting a scale homogeneity assumption within each data source (e.g., the nested logit trick) that varies between data sources. This paper extends the state of the art by incorporating data-source specific scale differences in scale heterogeneity setting across pooled RP and SP data set. An example of choice amongst RP and SP transport modes (including two ‘new’ SP modes) is used to obtain values of travel time savings that vary significantly between a model that accounts for scale heterogeneity differences within pooled RP and SP data, and the other where differences in scale heterogeneity is also accommodated between RP and SP data.     

The Mixed Logit model: The state of practice

The mixed logit model is considered to be the most promising state of the art discrete choice model currently available. Increasingly researchers and practitioners are estimating mixed logit models of various degrees of sophistication with mixtures of revealed preference and stated choice data. It is timely to review progress in model estimation since the learning curve is steep and the unwary are likely to fall into a chasm if not careful. These chasms are very deep indeed given the complexity of the mixed logit model. Although the theory is relatively clear, estimation and data issues are far from clear. Indeed there is a great deal of potential mis-inference consequent on trying to extract increased behavioural realism from data that are often not able to comply with the demands of mixed logit models. Possibly for the first time we now have an estimation method that requires extremely high quality data if the analyst wishes to take advantage of the extended behavioural capabilities of such models. This paper focuses on the new opportunities offered by mixed logit models and some issues to be aware of to avoid misuse of such advanced discrete choice methods by the practitioner.     

Understanding variability,habit and the effect of long period activity plan in modal choices: a day to day,week to week analysis on panel data

Understanding variability in individual behaviour is crucial for the comprehension of travel patterns and for the development and evaluation of planning policies. But, with only one notable exception, there are no studies on the intrinsic variability in the individual preferences for mode choices in absence of external changes in the transport infrastructures. This requires using continuous panel data. Few papers have studied mode choice with continuous panel data but mainly focused on the panel correlation. In this work we use a six-week travel diary survey to study the intrinsic variability in the individual preferences for mode choices, the effect of long period plans and habitual behaviour in the daily mode choices. Mixed logit models are estimated that account for the above effects as well as for systematic and random heterogeneity over individual preferences and responses. We also account for correlation over several time periods. Our results suggest that individual tastes for time and cost are fairly stable but there is a significant systematic and random heterogeneity around these mean values and in the preferences for the different alternatives. We found that there is a strong inertia effect in mode choice that increases with (or is reinforced by) the number of time the same tour is repeated. The sequence of mode choice made is influenced by the duration of the activity and the weekly structure of the activities     

An aggregate approach to model evacuee behavior for no-notice evacuation operations

This study proposes an aggregate approach to model evacuee behavior in the context of no-notice evacuation operations. It develops aggregate behavior models for evacuation decision and evacuation route choice to support information-based control for the real-time stage-based routing of individuals in the affected areas. The models employ the mixed logit structure to account for the heterogeneity across the evacuees. In addition, due to the subjectivity involved in the perception and interpretation of the ambient situation and the information received, relevant fuzzy logic variables are incorporated within the mixed logit structure to capture these characteristics. Evacuation can entail emergent behavioral processes as the problem is characterized by a potential threat from the extreme event, time pressure, and herding mentality. Simulation experiments are conducted for a hypothetical terror attack to analyze the models’ ability to capture the evacuation-related behavior at an aggregate level. The results illustrate the value of using a mixed logit structure when heterogeneity is pronounced. They further highlight the benefits of incorporating fuzzy logic to enhance the prediction accuracy in the presence of subjective and linguistic elements in the problem.     

Modelling mode choice for freight transport using advanced choice experiments

In this paper we use advanced choice modelling techniques to analyse demand for freight transport in a context of modal choice. To this end, a stated preference (SP) survey was conducted in order to estimate freight shipper preferences for the main attributes that define the service offered by the different transport modes. From a methodological point of view, we focus on two critical issues in the construction of efficient choice experiments. Firstly, in obtaining good quality prior information about the parameters; and secondly, in the improved quality of the experimental data by tailoring a specific efficient design for every respondent in the sample.With these data, different mixed logit models incorporating panel correlation effects and accounting for systematic and random taste heterogeneity are estimated. For the best model specification we obtain the willingness to pay for improving the level of service and the elasticity of the choice probabilities for the different attributes. Our model provide interesting results that can be used to analyse the potential diversion of traffic from road (the current option) to alternative modes, rail or maritime, as well as to help in the obtaining of the modal distribution of commercial traffic between Spain and the European Union, currently passing through the Pyrenees.     

Active school travel: homogeneity or heterogeneity? That is the question

ABSTRACT

To explain and predict active school travel (AST), most studies have not investigated to what extent considering taste heterogeneity is an important influence on AST share. The main aim of the present study was to evaluate whether considering unobserved taste heterogeneity through mixed logit models – including random coefficient and random coefficient analysis (RCA) – materially improves/influences the AST prediction compared to a simpler model – the multinomial logit (MNL) model. The database comprises 735 valid observations. The results show that, with a 10% increase in perceived walking time to school, the MNL model predicts that the AST share would decrease by 7.8% (from 18.9% to 17.4%) while the RCA model predicts that it would decrease by 8.5% (from 18.9% to 17.3%). Thus, the expected share of AST is overestimated by MNL by one-tenth of a percentage point. Although there might be random taste variations around perceived distance to school, it seems the other important policy-sensitive variables, such as safety perception, homogeneously impacts on the AST share across households with different socioeconomic and built environment characteristics. Our empirical assessment suggests that considering taste heterogeneity does not necessarily improve the accuracy of analysis for the aggregate share of the AST concerning policy-sensitive variables.     

Accounting for systematic heterogeneity across car commuters in response to multiple TDM policies: case study of Tehran

Modeling commuters’ choice behavior in response to transportation demand management (TDM) helps in predicting the consequences of TDM policies. Although research looking at choice behavior has evolved to investigate preference heterogeneity in response to factors influencing mode choice, as far as we know, no study has considered taste variation across commuters in response to multiple TDM policies. This paper investigates the presence of systematic preference heterogeneity across commuters, in response to the TDM policies that can be explained by their socio-economic or commuting-related characteristics. Analysis is based on results of a stated preference survey developed using a Design of Experiments approach. Five policies were assessed in order to study the impact they had on how commuters chose their mode of transportation. These include increasing parking cost, increasing fuel cost, implementing cordon pricing, reducing transit time and improving access to transit facilities. For the sake of assessing both systematic and random preference heterogeneity across car commuters, a form of the Mixed Multinomial Logit (MMNL) model that identifies sources of heterogeneity and consequently makes the choice models less restrictive in considering both systematic and random preference variation across individuals was developed. The sample includes 366 individuals who regularly commute to their workplace in the city center of Tehran, Iran. The likelihood function value of this model shows a significant improvement compared to the base MNL model, using the same variables. The MMNL model shows that taste variation across the studied commuters results in differences in influences estimated for three policies: increasing parking cost, reducing transit time and improving access to transit. The analysis examines several distributions for random parameters to test the impacts of restricting distributions to allow for only normality. The results confirm the potential to improve model fit with alternative distributions.     

A joint model for vehicle type and fuel type choice: evidence from a cross-nested logit study

In the face of growing concerns about greenhouse gas emissions, there is increasing interest in forecasting the likely demand for alternative fuel vehicles. This paper presents an analysis carried out on stated preference survey data on California consumer responses to a joint vehicle type choice and fuel type choice experiment. Our study recognises the fact that this choice process potentially involves high correlations that an analyst may not be able to adequately represent in the modelled utility components. We further hypothesise that a cross-nested logit structure can capture more of the correlation patterns than the standard nested logit model structure in such a multi-dimensional choice process. Our empirical analysis and a brief forecasting exercise produce evidence to support these assertions. The implications of these findings extend beyond the context of the demand for alternative fuel vehicles to the analysis of multi-dimensional choice processes in general. Finally, an extension verifies that further gains can be made by using mixed GEV structures, allowing for random heterogeneity in addition to the flexible correlation structures.     

Preference Heterogeneity and Willingness to Pay for Travel Time Savings

We examined different model specifications to detect the presence of preference heterogeneity in a mode choice context. The specification that worked best allows for both systematic and random variations in tastes, with parameters obtained at the individual level using Bayesian methods. Subjective values of travel time (SVT) and expected individual compensated variation were derived and aggregated to obtain measures of social welfare. Results suggest that the benefit measures, both at the individual and at the social level, are sensitive to preference heterogeneity assumptions. SVT and welfare changes derived from travel time reductions could be underestimated if the traditional assumption of taste homogeneity is made (we detected differences up to 30% in both types of measures). We also obtained an empirical value for the error made when evaluating changes in social welfare using an approximation of the expected individual compensated variation (expressed as a function of individual SVT) rather than its exact expression.     

Are consumers willing to pay to let cars drive for them? Analyzing response to autonomous vehicles

Autonomous vehicles use sensing and communication technologies to navigate safely and efficiently with little or no input from the driver. These driverless technologies will create an unprecedented revolution in how people move, and policymakers will need appropriate tools to plan for and analyze the large impacts of novel navigation systems. In this paper we derive semiparametric estimates of the willingness to pay for automation. We use data from a nationwide online panel of 1260 individuals who answered a vehicle-purchase discrete choice experiment focused on energy efficiency and autonomous features. Several models were estimated with the choice microdata, including a conditional logit with deterministic consumer heterogeneity, a parametric random parameter logit, and a semiparametric random parameter logit. We draw three key results from our analysis. First, we find that the average household is willing to pay a significant amount for automation: about $3500 for partial automation and $4900 for full automation. Second, we estimate substantial heterogeneity in preferences for automation, where a significant share of the sample is willing to pay above $10,000 for full automation technology while many are not willing to pay any positive amount for the technology. Third, our semiparametric random parameter logit estimates suggest that the demand for automation is split approximately evenly between high, modest and no demand, highlighting the importance of modeling flexible preferences for emerging vehicle technology.     

Valuing of attributes influencing the attractiveness of suburban train service in Mumbai city: A stated preference approach

The paper presents valuing of qualitative and quantitative travel attributes influencing the attractiveness of suburban train service in Mumbai city, India. A stated preference experiment is designed to capture the data of sub-urban train mode choice behavior. The behavioral data are analyzed using different modeling techniques such as multinomial logit (MNL) and mixed logit (ML). In ML model, the random parameters are assumed to follow constrained triangular distribution, where mean equals its spread. The decomposition of preference heterogeneity around the mean estimate of random parameter is also investigated using ML model. The study shows the influence of headway time and train ride time associated with a particular crowding level (expressed in density of standing passengers/m²) in choosing the sub-urban train mode by calculating their willingness-to-pay (WTP) values and highlights the importance of WTP for addressing policy issues in the reduction of in-vehicle crowding level. The present study documents new findings of the effect of crowding level on train ride time in the context of a developing country and suggests some important directions for future suburban train transport crowding valuation research.     

The Signs of the Times: Imposing a Globally Signed Condition on Willingness to Pay Distributions

A feature of recent developments in choice models that enable estimation of the distribution of willingness to pay (WTP) is that the sign of the distribution can change over the range. Behaviourally this often makes little sense for attributes such as travel time on non-discretionary travel, despite a growing recognition of positive utility over some travel time ranges. This can in part be attributed to the analytical distribution that is selected (except the cumbersome lognormal), many of which are unconstrained over the full range. Although a number of analysts have imposed constraints on various distributions for random parameters that can satisfy the single-sign condition, these restrictions are, with rare exception, only satisfied for the mean and the standard deviation estimates of a random parameter. When heterogeneity around the mean and/or heteroscedasticity around the standard deviation is allowed for, however, the constraint condition is often not satisfied. Given the popularity of distributions other than the lognormal, in order to satisfy the sign condition under the most general form of parameterisation, we need to impose a global sign condition. In this paper we show how this might be achieved in the context of the valuation of travel time savings for car commuters choosing amongst an offered set of route-specific travel times and costs. We illustrate the impact of the constraint under a globally constrained Rayleigh distribution for total travel time parameterisation, contrasting the evidence with a multinomial logit model and a range of other distributional assumptions within the mixed logit framework. Discussions with Bill Greene, John Rose, Ken Train and especially Juan de Dios Ortuzar have been invaluable as have the comments of referees.     

Embedding risk attitude and decision weights in non-linear logit to accommodate time variability in the value of expected travel time savings

In recent years we have seen important extensions of logit models in behavioural research such as incorporation of preference and scale heterogeneity, attribute processing heuristics, and estimation of willingness to pay (WTP) in WTP space. With rare exception, however, a non-linear treatment of the parameter set to allow for behavioural reality, such as embedded risk attitude and perceptual conditioning of occurrence probabilities attached to specific attributes, is absent. This is especially relevant to the recent focus in travel behaviour research on identifying the willingness to pay for reduced travel time variability, which is the source of estimates of the value of trip reliability that has been shown to take on an increasingly important role in project appraisal. This paper incorporates, in a generalised non-linear (in parameters) logit model, alternative functional forms for perceptual conditioning (known as probability weighting) and risk attitude in the utility function to account for travel time variability, and then derives an empirical estimate of the willingness to pay for trip time variability-embedded travel time savings as an alternative to separate estimates of time savings and trip time reliability. We illustrate the richness of the approach using a stated choice data set for commuter choice between unlabelled attribute packages. Statistically significant risk attitude parameters and parameters underlying decision weights are estimated for multinomial logit and mixed multinomial logit models, along with values of expected travel time savings.     

The valuation of commuter travel time savings for car drivers: evaluating alternative model specifications

The empirical valuation of travel time savings is a derivative of the ratio of parameter estimates in a discrete choice model. The most common formulation (multinomial logit) imposes strong restrictions on the profile of the unobserved influences on choice as represented by the random component of a preference function. As we progress our ability to relax these restrictions we open up opportunities to benchmark the values derived from simple (albeit relatively restrictive) models. In this paper we contrast the values of travel time savings derived from multinomial logit and alternative specifications of mixed (or random parameter) logit models. The empirical setting is urban car commuting in six locations in New Zealand. The evidence suggests that less restrictive choice model specifications tend to produce higher estimates of values of time savings compared to the multinomial logit model; however the degree of under-estimation of multinomial logit remains quite variable, depending on the context.     

Microscopic decision model for pedestrian route choice at signalized crosswalks

In this paper, two‐tier mathematical models were developed to simulate the microscopic pedestrian decision‐making process of route choice at signalized crosswalks. In the first tier, a discrete choice model was proposed to predict the choices of walking direction. In the second tier, an exponential model was calibrated to determine the step size in the chosen direction. First, a utility function was defined in the first‐tier model to describe the change of utility in response to deviation from a pedestrian's target direction and the conflicting effects of neighboring pedestrians. A mixed logit model was adopted to estimate the effects of the explanatory variables on the pedestrians' decisions. Compared with the standard multinomial logit model, it was shown that the mixed logit model could accommodate the heterogeneity. The repeated observations for each pedestrian were grouped as panel data to ensure that the parameters remained constant for individual pedestrians but varied among the pedestrians. The mixed logit model with panel data was found to effectively address inter‐pedestrian heterogeneity and resulted in a better fit than the standard multinomial logit model. Second, an exponential model in the second tier was proposed to further determine the step size of individual pedestrians in the chosen direction; it indicates the change in walking speed in response to the presence of other pedestrians. Finally, validation was conducted on an independent set of observation data in Hong Kong. The pedestrians' routes and destinations were predicted with the two‐tier models. Compared with the tracked trajectories, the average error between the predicted destinations and the observed destinations was within an acceptable margin. Copyright © 2016 John Wiley & Sons, Ltd.     

Modeling heterogeneous parking choice behavior on university campuses

Studies on campus parking indicate more severe problems and a wider range of characteristics than commercial parking because of limited parking places, special conditions, specific policies and enclosed space on university campuses. Heterogeneous characteristics are usually ignored in analyses of campus parking behavior. In this paper, a mixed logit model is applied to analyze parking choice behavior on a campus using data collected from a stated-preference survey of Tongji University, Shanghai, China. The heterogeneity of individuals with various sociodemographic characteristics is evaluated by interaction terms and random parameters. Comparison between the proposed approach and the conditional logit model shows that the results of the mixed logit model are more interpretable because they are not limited by the independence from irrelevant alternatives assumption. Key factors that have considerable effects on campus parking choices are identified and analyzed. Important regularities are also concluded from elasticity analyses. Finally, the campus is divided into two areas according to the walking distance to a new parking lot, and the modeling results show that area-specific policies should be established because the two areas have quite distinct parking choice features.     

The accuracy of the multinomial logit model as an approximation to the multinomial probit model of travel demand

The multinomial probit model of travel demand is considerably more general but much less tractable than the better-known multinomial logit model. In an effort to determine the effects of using the relatively simple logit model in situations where the assumptions of probit modeling are satisfied but those of logit modeling are not, the accuracy of the multinomial logit model as an approximation to a variety of three-alternative probit models has been evaluated. Multinomial logit can give highly erroneous estimates of the choice probabilities of multinomial probit models. However, logit models appear to give asymptotically accurate estimates of the ratios of the coefficients of the systematic components of probit utility functions, even when the logit choice probabilities differ greatly from the probit ones. Large estimation data sets are not necessarily needed to enable likelihood ratio tests to distinguish three-alternative probit models from logit models that give seriously erroneous estimates of the probit choice probabilities. Inclusion of alternative-specific dummy variables in logit utility functions cannot be relied upon to reduce significantly the errors of logit approximations to the choice probabilities of probit models whose utility functions do not contain the dummies.     

A utility maximizing model of the demand for multi-destination non-work travel

The multinomial probit model of travel demand is considerably more general but much less tractable than the better-known multinomial logit model. In an effort to determine the effects of using the relatively simple logit model in situations where the assumptions of probit modeling are satisfied but those of logit modeling are not, the accuracy of the multinomial logit model as an approximation to a variety of three-alternative probit models has been evaluated. Multinomial logit can give highly erroneous estimates of the choice probabilities of multinomial probit models. However, logit models appear to give asymptotically accurate estimates of the ratios of the coefficients of the systematic components of probit utility functions, even when the logit choice probabilities differ greatly from the probit ones. Large estimation data sets are not necessarily needed to enable likelihood ratio tests to distinguish three-alternative probit models from logit models that give seriously erroneous estimates of the probit choice probabilities. Inclusion of alternative-specific dummy variables in logit utility functions cannot be relied upon to reduce significantly the errors of logit approximations to the choice probabilities of probit models whose utility functions do not contain the dummies.