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.     

Trip chaining as a barrier to the propensity to use public transport

Trip chaining is a growing phenomenon in travel and activity behaviour. Individuals increasingly seek out opportunities to minimise the amount of travel required as part of activity fulfilment, given the competing demands on time budgets and their valuation of travel time savings. This search for ways of fulfilling (more) activities with less travel input has produced a number of responses, one of which is trip chaining. A particularly important policy implication of trip chaining is the potential barrier it creates in attracting car users to switch to public transport. This paper seeks to improve our understanding of trip chaining as a barrier to public transport use. A series of discrete choice models are estimated to identify the role that socio-economic and demographic characteristics of households have on the propensity to undertake trip chains of varying degrees of simplicity/complexity that involve use of the car or public transport with an embedded commuting or non-commuting primary purpose. Multinomial logit, nested logit and random parameter logit models are developed and contrasted to establish the gains in relaxing the strict conditions of the multinomial logit model.     

Experimental design influences on stated choice outputs: An empirical study in air travel choice

Discrete choice experiments are conducted in the transport field to obtain data for investigating travel behaviour and derived measures such as the value of travel time savings. The multinomial logit (MNL) and other more advanced discrete choice models (e.g., the mixed MNL model) have often been estimated on data from stated choice experiments and applied for planning and policy purposes. Determining efficient underlying experimental designs for these studies has become an increasingly important stream of research, in which the objective is to generate stated choice tasks that maximize the collected information, yielding more reliable parameter estimates. These theoretical advances have not been rigorously tested in practice, such that claims on whether the theoretical efficiency gains translate into practice cannot be made. Using an extensive empirical study of air travel choice behaviour, this paper presents for the first time results of different stated choice experimental design approaches, in which respective estimation results are compared. We show that D-efficient designs keep their promise in lowering standard errors in estimating, thereby requiring smaller sample sizes, ceteris paribus, compared to a more traditional orthogonal design. The parameter estimates found using an orthogonal design or an efficient design turn out to be statistically different in several cases, mainly attributed to more or less dominant alternatives existing in the orthogonal design. Furthermore, we found that small designs with a limited number of choice tasks performs just as good (or even better) than a large design. Finally, we show that theoretically predicted sample sizes using the so-called S-estimates provide a good lower bound. This paper will enable practitioners in better understanding the potential benefits of efficient designs, and enables policy makers to make decisions based on more reliable parameter estimates.     

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.     

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.     

A nested logit model of automobile holdings for one vehicle households

This paper presents a model of automobile choice by single vehicle households. This effort is distinguished from previous disaggregate automobile holdings models primarily by the use of the nested logit model rather than the more restrictive multinomial logit model. We present a 2-step estimation technique that provides consistent and asymptotically efficient parameter estimates, yet is tractable for very large choice sets. Using disaggregate data on 237 one-vehicle households we estimate the unknown parameters on an automobile choice model containing 785 individual makes, models and vintages of passenger vehicles.     

A theoretical and empirical model of trip chaining behavior

This paper addresses the theoretical and empirical issues involved in modeling complex travel patterns. Existing models have the shortcoming of not representing the interdependencies among trip links in trip chains with multiple non-home stops. A theoretical model based on utility theory and explicitly accounting for the trade-offs involved in the choice of multiple-stop chains is developed. Using this theoretical model, utility maximizing conditions for a household's choice of a daily travel pattern are derived. The optimum travel pattern is described in terms of the number of chairs (tours) traveled on a given day and in terms of the number of stops (sojourns) made on each of those chains. For a given household, the form of the optimum pattern is a function of the transportation expenditures (time, cost) required to reach potential destinations. Constraints on the conditions of optimality due to the limited and discrete nature of travel pattern alternatives are also considered. Parameters of the general utility function were estimated empirically using actual travel data derived from a home interview survey taken in Washington, D.C. The multinomial logit model is used to relate utility scores for the alternative travel patterns to choice probabilities. The resulting parameter estimates agree with theoretical expectations and with empirical results obtained in other studies. In order to demonstrate the empirical and theoretical implications of the model, forecasts for various transportation policies (e.g., gasoline price increases, transit fare reductions), as made by this model and by other less complex models, are compared. The results of these comparisons indicate the need for expanding the scope of existing travel forecasting models to explicit considerations of trip chaining behavior.     

The multinomial logit model revisited: A semi-parametric approach in discrete choice analysis

The multinomial logit model in discrete choice analysis is widely used in transport research. It has long been known that the Gumbel distribution forms the basis of the multinomial logit model. Although the Gumbel distribution is a good approximation in some applications such as route choice problems, it is chosen mainly for mathematical convenience. This can be restrictive in many other scenarios in practice. In this paper we show that the assumption of the Gumbel distribution can be substantially relaxed to include a large class of distributions that is stable with respect to the minimum operation. The distributions in the class allow heteroscedastic variances. We then seek a transformation that stabilizes the heteroscedastic variances. We show that this leads to a semi-parametric choice model which links the linear combination of travel-related attributes to the choice probabilities via an unknown sensitivity function. This sensitivity function reflects the degree of travelers’ sensitivity to the changes in the combined travel cost. The estimation of the semi-parametric choice model is also investigated and empirical studies are used to illustrate the developed method.     

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.     

Household trip generation choice—Alternative empirical approaches

Trip generation models have generally received less attention than other aspects of travel decision making. This article presents some explorations into the structure of trip decisions for shopping, using data from weekly shopping diaries. The paper compares alternative formulations of the naturally-ordered choice model used by Sheffi (1979) to avoid the problems inherent in multinomial logit models. Firstly, imposing cross-alternative restrictions on some of the coefficient values is termed the constrained model. Secondly, when no such restrictions are imposed, the model decomposes into a sequence of binary models, and this is termed the unconstrained model, which can be used to test the validity of the restrictions. The variables used include both shopping expenditures and locational factors, both of which are found to play a key role in shopping travel decisions, as well as more conventional socio-economic variables. A clearer understanding of the role of car ownership in travel decisions is obtained.     

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.     

Multidimensional choice model transferability

Transferability studies have focused on the component models of the conventional four-step urban travel forecasting model system. This study extends previous analyses by examining the transferability of models describing multidimensional travel and related choices. In particular, we examine the hypothesis that joint and sequential choice models are equally transferable against the alternative hypotheses that either of the model types is more transferable. Measures of goodness of fit and transfer effectiveness are formulated for sequential choice models to provide a consistent comparison between the joint and sequential models. An empirical analysis is undertaken in the context of joint (multinomial logit) and sequential (nested logit) models of automobile ownership and mode choice to work. This study finds little difference between the transferability of these joint and sequential models. However, this conclusion appears to be dependent on the similarity of the estimation results for the joint and sequential models in this case. These results suggest a need for additional testing in other empirical contexts to identify the relative transferability of joint versus sequential models when the estimation results are distinct.     

Development of commuter and non-commuter mode choice models for the assessment of new public transport infrastructure projects: A case study

This paper uses state of the art stated choice designs to parameterise modal choice models for commuting and non-commuting travel futures in the presence of new public transport infrastructure (variations of new heavy rail, light rail and dedicated busway systems). D-optimal choice experiments are developed for a set of labelled modal alternatives in which respondents establish a reference benchmark based on the existing service levels (for access, linehaul and egress trip legs) which is used in a computer aided personal interview instrument to generate future scenarios of service levels for current and prospective new modals options. We show that a fully integrated stated choice experiment provides all the information required to obtain behaviourally relevant parameter estimates (within a nested logit framework) for all but the mode-specific constants (MSCs). The MSCs can be calibrated for the current modes within a network model setting, giving the transport planner an appropriate model for predicting the patronage potential for proposed new public transport infrastructure services. A useful by-product is a new set of behavioural values of travel time savings for access, egress, linehaul and wait times.     

Development of railway station choice models to improve the representation of station catchments in rail demand models

ABSTRACT

This paper describes the development of railway station choice models suitable for defining probabilistic station catchments. These catchments can then be incorporated into the aggregate demand models typically used to forecast demand for new rail stations. Revealed preference passenger survey data obtained from the Welsh and Scottish Governments was used for model calibration. Techniques were developed to identify trip origins and destinations from incomplete address information and to automatically validate reported trips. A bespoke trip planner was used to derive mode-specific station access variables and train leg measures. The results from a number of multinomial logit and random parameter (mixed) logit models are presented and their predictive performance assessed. The models were found to have substantially superior predictive accuracy compared to the base model (which assumes the nearest station has a probability of one), indicating that their incorporation into passenger demand forecasting methods has the potential to significantly improve model predictive performance.     

Determinants of ground transport modal choice in long-distance trips in Spain

In transport economics, modeling modal choice is a fundamental key for policy makers trying to improve the sustainability of transportation systems. However, existing empirical literature has focused on short-distance travel within urban systems. This paper contributes to the limited number of investigations on mode choice in medium- and long-distance travel. The main objective of this research is to study the impacts of socio-demographic and economic variables, land-use features and trip attributes on long-distance travel mode choice. Using data from 2007 Spanish National Mobility Survey we apply a multilevel multinomial logit model that accounts for the potential problem of spatial heterogeneity in order to explain long-distance travel mode choice. This approach permits us to compute how the probability of choosing among private car, bus and train varies depending on the traveler spatial location at regional level. Results indicate that travelers characteristics, trip features, cost of usage of transport modes and geographical variables have significant impacts on long-distance mode choice.     

Sampling of alternatives for route choice modeling

This paper presents a new paradigm for choice set generation in the context of route choice model estimation. We assume that the choice sets contain all paths connecting each origin–destination pair. Although this is behaviorally questionable, we make this assumption in order to avoid bias in the econometric model. These sets are in general impossible to generate explicitly. Therefore, we propose an importance sampling approach to generate subsets of paths suitable for model estimation. Using only a subset of alternatives requires the path utilities to be corrected according to the sampling protocol in order to obtain unbiased parameter estimates. We derive such a sampling correction for the proposed algorithm.Estimating models based on samples of alternatives is straightforward for some types of models, in particular the multinomial logit (MNL) model. In order to apply MNL for route choice, the utilities should also be corrected to account for the correlation using, for instance, a path size (PS) formulation. We argue that the PS attribute should be computed based on the full choice set. Again, this is not feasible in general, and we propose a new version of the PS attribute derived from the sampling protocol, called Expanded PS.Numerical results based on synthetic data show that models including a sampling correction are remarkably better than the ones that do not. Moreover, the Expanded PS shows good results and outperforms models with the original PS formulation.     

A simulation evaluation of the maximum approximate composite marginal likelihood (MACML) estimator for mixed multinomial probit models

This paper evaluates the ability of the maximum approximate composite marginal likelihood (MACML) estimation approach to recover parameters from finite samples in mixed cross-sectional and panel multinomial probit models. Comparisons with the maximum simulated likelihood (MSL) estimation approach are also undertaken. The results indicate that the MACML approach recovers parameters much more accurately than the MSL approach in all model structures and covariance specifications. The MACML inference approach also estimates the parameters efficiently, with the asymptotic standard errors being, in general, only a small proportion of the true values. As importantly, the MACML inference approach takes only a very small fraction of the time needed for MSL estimation. In particular, the results suggest that, for the case of five random coefficients, the MACML approach is about 50 times faster than the MSL for the cross-sectional random coefficients case, about 15 times faster than the MSL for the panel inter-individual random coefficients case, and about 350 times or more faster than the MSL for the panel intra- and inter-individual random coefficients case. As the number of alternatives in the unordered-response model increases, one can expect even higher computational efficiency factors for the MACML over the MSL approach. Further, as should be evident in the panel intra- and inter-individual random coefficients case, the MSL is all but practically infeasible when the mixing structure leads to an explosion in the dimensionality of integration in the likelihood function, but these situations are handled with ease in the MACML approach. It is hoped that the MACML procedure will spawn empirical research into rich model specifications within the context of unordered multinomial choice modeling, including autoregressive random coefficients, dynamics in coefficients, space-time effects, and spatial/social interactions.     

Confidence intervals of willingness-to-pay for random coefficient logit models

Random coefficient logit (RCL) models containing random parameters are increasingly used for modelling travel choices. Willingness-to-pay (WTP) measures, such as the value of travel time savings (VTTS) are, in the case of RCL models estimated in preference space, ratios of random parameters. In this paper we apply the Delta method to compute the confidence intervals of such WTP measures, taking into account the variance–covariance matrix of the estimates of the distributional parameters. The same Delta method can be applied when the model is estimated in WTP space. Compared to simulation methods such as proposed by Krinsky and Robb, the Delta method is able to avoid most of the simulations by deriving partly analytical expressions for the standard errors. Examples of such computations are shown for different combinations of random distributions.     

The value of travel time and reliability-evidence from a stated preference survey and actual usage

This research examined travel behavior of Managed Lane (ML) users to better understand the value travelers place on travel time savings and travel time reliability. We also highlight the importance of survey design techniques. These objectives were accomplished through a stated preference survey of Houston’s Katy Freeway travelers. Three stated choice experiment survey design techniques were tested in this study: Bayesian (D_b) efficient, random level attribute generation, and an adaptive random approach. Mixed logit models were developed from responses using each of those designs. The value of travel time savings (VTTSs) estimates do vary across the design strategies, with the VTTS estimates based on the D_b-efficient design being approximately half the estimates from the other two designs. However, among the three design strategies, the value of travel time reliability (VOR) was only significant in the D_b-efficient design.The estimated VTTS from actual Katy Freeway usage (as measured using actual tolls paid and travel time saved on the managed lanes) is $51/h. This likely also includes any value that travelers place on travel time reliability. In comparison, our combined estimate of VTTS and VOR based on the SP survey (D_b-efficient design) was $50/h, which is remarkably close to the estimate from the actual usage data. Based on our dataset, the D_b-efficient design technique proved superior to the other two techniques. Finally, this research also supports the importance of including both travel time and travel time reliability parameters when estimating the willingness to pay for, and therefore benefits derived from, ML travel.