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.     

Identification and diagnosis of specification errors in the multinomial logit model

Multinomial logit models of travel demand are subject to a variety of specification errors that can severely affect the accuracy the models' forecasts. This paper describes some of the important sources of specification errors in logit models, presents numerical illustrations of the errors' effects on forecasts of travel behavior, and evaluates the ability of several informal and formal statistical procedures to identify and diagnose the errors. Among the tests considered, the most powerful are a test against a probit model and the McFadden, Tye and Train (1976) test based on the universal logit method. The least powerful tests are a test based on extrapolating the logit model and informal tests based on examination of the signs, t-statistics and ratios of the estimated parameters of the logit function.     

Residential mobility and location choice: a nested logit model with sampling of alternatives

Residential mobility and relocation choice are essential parts of integrated transportation and land use models. These decision processes have been examined and modeled individually to a great extent but there remain gaps in the literature on the underlying behaviors that connect them. Households may partly base their decision to move from or stay at a current location on the price and quality of the available alternatives. Conversely, households that are on the market for a new location may evaluate housing choices relative to their previous residence. How and the degree to which these decisions relate to each other are, however, not completely understood. This research is intended to contribute to the body of empirical evidence that will help answer these questions. It is hypothesized that residential mobility and location choice are related household decisions that can be modeled together using a two-tier hierarchical structure. This paper presents a novel nested logit (NL) model with sampling of alternatives and a proposed procedure for sampling bias correction. The model was estimated using full-information maximum likelihood estimation methods. The results confirm the applicability of this NL model and support similar findings from other empirical studies in the residential mobility and location choice literature.     

Understanding public response to a congestion charge: A random-effects ordered logit approach

Public acceptance is consistently listed as having an enormous impact on the implementation and success of a congestion charge scheme. This paper investigates public acceptance of such a scheme in Australia. Surveys were conducted in Brisbane and Melbourne, the two fastest growing Australian cities. Using an ordered logit modeling approach, the survey data including stated preferences were analyzed to pinpoint the important factors influencing people’s attitudes to a congestion charge and, in turn, to their transport mode choices. To accommodate the nature of, and to account for the resulting heterogeneity of the panel data, random effects were considered in the models. As expected, this study found that the amount of the congestion charge and the financial benefits of implementing it have a significant influence on respondents’ support for the charge and on the likelihood of their taking a bus to city areas. However, respondents’ current primary transport mode for travelling to the city areas has a more pronounced impact. Meanwhile, respondents’ perceptions of the congestion charge’s role in protecting the environment by reducing vehicle emissions, and of the extent to which the charge would mean that they travelled less frequently to the city for shopping or entertainment, also have a significant impact on their level of support for its implementation. We also found and explained notable differences across two cities. Finally, findings from this study have been fully discussed in relation to the literature.     

Constrained nested logit model: formulation and estimation

A model of traveller behaviour should recognise the exogenous and endogenous factors that limit the choice set of users. These factors impose constraints on the decision maker, which constraints may be considered implicitly, as soft constraints imposing thresholds on the perception of changes in attribute values, or explicitly as hard constraints. The purpose of this paper is twofold: (1) To present a constrained nested logit-type choice model to cope with hard constraints. This model is derived from the entropy-maximizing framework. (2) To describe a general framework to deal with (dynamic) non-linear utilities. This approach is based on Reproducing Kernel Hilbert Spaces. The resulting model allows the dynamic aspect and the constraints on the choice process to be represented simultaneously. A novel estimation procedure is introduced in which the utilities are viewed as the parameters of the proposed model instead of attribute weights as in the classical linear models. A discussion on over-specification of the proposed model is presented. This model is applied to a synthetic test problem and to a railway service choice problem in which users choose a service depending on the timetable, ticket price, travel time and seat availability (which imposes capacity constraints). Results show (1) the relevance of incorporating constraints into the choice models, (2) that the constrained models appear to be a better fit than the counterpart unconstrained choice models; and (3) the viability of the approach, in a real case study of railway services on the Madrid–Seville corridor (Spain).     

Capturing tour mode and activity choice interdependencies: A co-evolutionary logit modelling approach

It is generally assumed that the choice of transport mode and the choice of including intermediate activities on a work tour are interrelated, but little is known about the nature of the causal relationship. To shed light on this, this paper addresses the question of whether transport mode choice is dependent on the activity choice or vice-versa. A new methodology, referred to as the co-evolutionary approach, is combined with a set of MNL models, one for each choice facet involved, to derive an indication of the order of decisions on an individual level. The models are estimated based on the work tours of a large sample of individuals in the Netherlands. The results suggest that there is substantial variation in the order of the transport mode and activity decisions. However, in the majority of cases the activity decision is made before the mode decision, suggesting that the transport mode and, in particular, the choice between car and public transport is most often ‘adjusted’ to the choice of trip chaining rather than the other way round.     

Modeling residential location choice in relation to housing location and road tolls on congested urban highway networks

We present a reformulation of the residential location submodel of the Integrated Model of Residential and Employment Location as a network equilibrium problem, thereby making travel costs by auto endogenous. The location of housing supply is examined as a welfare maximization problem for both user-optimal and system-optimal travel costs using concepts of bilevel programming. Finally, we briefly discuss how the employment submodel can be reformulated, and the entire model solved as a variational inequality problem.     

Aggregation with multinomial probit and estimation of disaggregate models with aggregate data: A new methodological approach

This paper describes an analytic aggregation procedure for disaggregate demand models similar to the one proposed in earlier publications by Westin (1974) and McFadden and Reid (1975). The technique, which uses a multivariate normal approximation for the distribution of the vector of attributes, is based on the multinomial profit algorithm proposed by Daganzo, Bouthelier and Sheffi (1977) and can be applied to an arbitrary number of alternatives. The procedure is computationally so efficient that it enables us to calibrate disaggregate models with aggregate data by maximum likelihood using the same or slightly modified codes developed for disaggregated data. The paper also contains a small scale numerical example intended to illustrate the important highlights of the aggregation-estimation problem.     

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.     

Commodity flow estimation for a metropolitan scale freight modeling system: supplier selection considering distribution channel using an error component logit mixture model

Freight forecasting models have been significantly improved in recent years, especially in the field of goods vehicle behavior modeling. On the other hand, the improvements to commodity flow modeling, which provide inputs for goods vehicle simulations, were limited. Contributing to this component in urban freight modeling systems, we propose an error component logit mixture model for matching a receiver to a supplier that considers two-layers in supplier selection: distribution channels and specific suppliers. The distribution channel is an important element in freight modeling, as the type of distribution channel is relevant to various aspects of shipments and vehicle trips. The model is estimated using the data from the Tokyo Metropolitan Freight Survey. We demonstrate how typical establishment survey data (i.e. establishment and outbound shipment records) can be used to develop the model. The model captures the correlation structure of potential suppliers defined by business function and provides insights on the differences in the supplier choice by distribution channel. The reproducibility tests confirm the validity of the proposed approach, which is currently integrated into a metropolitan-scale agent-based freight modeling system, for practical use.

     

Dynamic location: an iconic model to synchronize temporal and spatial transportation data

This paper describes a model to synchronize the management and query of temporal and spatially referenced transportation data in geographic information systems (GIS). The model employs a method referred to as dynamic location, which facilitates spatial intersect queries from geographic shapes without the use of topological relationships. This is the inverse of how dynamic segmentation works in GIS. In contrast to dynamic segmentation, dynamic location stores geometry as an object within a single database field. This is an efficient, precise iconic model superseding the need for data decomposition into a complex set of tables. As an object model, the dynamic location process lends itself to high performance in an Internet, data-intensive, enterprise environment. Linear events are stored as {x, y} features, and not referenced to any route system. Route systems are built from {x, y, m} values (m for measure) and serve as number lines for mathematical operations. Any {x, y} object can then be referenced to either the Cartesian grid or any selected number line. This method offers the benefits of linear referencing, while making full use of a stable geodetic datum. Combinations of any {x, y} events may be placed over any {x, y, m} number line (route) and an intersect determined by looking through stacked {x, y, m} vertices of the coincident shapes. Since both geometry and shape reside in the same record, the use of “begin” and “end” dates facilitates full spatial and temporal version control. From a business process perspective, this creates a spatially enabled database, pulling GIS business functions back into the information technology mainstream.     

Effects of task complexity and time pressure on activity-travel choices: heteroscedastic logit model and activity-travel simulator experiment

This paper derives, estimates and applies a discrete choice model of activity-travel behaviour that accommodates potential effects of task complexity and time pressure on decision-making. To the best of our knowledge, this is the first time that both factors (task complexity and time pressure) are jointly captured in a discrete choice model. More specifically, our heteroscedastic logit model captures potential impacts of task complexity and time pressure through the scale of the utility of activity-travel options. We collect data using a novel activity-travel simulator experiment that has been specifically designed with the aim of testing our model. Results are in line with expectations, in that higher levels of task complexity and time pressure are found to result in a smaller scale of utility. In other words, higher levels of task complexity and time pressure lead to more random choice behaviour and as a consequence to less pronounced differences in choice probabilities between alternatives. An empirical illustration suggests that choice probability-differences between models that do and those that do not capture these effects, can be very substantial; this in turn suggests that failing to capture the effects of task complexity and time pressure in discrete choice models of activity travel decision-making might lead to serious bias in forecasts of the effects of transport policies.     

Values,attitudes and travel behavior: a hierarchical latent variable mixed logit model of travel mode choice

Values lie at the heart of an individual’s belief system, serving as prototypes from which attitudes and behaviors are subsequently manufactured. Attitudes and behaviors may evolve over time, but values represent a set of more enduring beliefs. This study examines the influence of values on travel mode choice behavior. It is argued that personal values influence individual attitudes towards different alternative attributes, which in turn impact modal choices. Using data from a sample of 519 German commuters drawn from a consumer panel, the study estimates an integrated choice and latent variable model of travel mode choice that allows for hierarchical relationships between the latent variables and flexible substitution patterns across the modal alternatives. Results from the empirical application support the value-attitude-behavior hierarchical model of cognition, and provide insights to planners and policy-makers on how better to sell public transit as a means of travel.     

Enhanced traffic information dissemination to facilitate toll road utilization: a nested logit model of a stated preference survey in Texas

This paper investigates the effects of the provision of traffic information on toll road usage based on a stated preference survey conducted in central Texas. Although many researchers have studied congestion pricing and traffic information dissemination extensively, most of them focused on the effects that these instruments individually produce on transportation system performance. Few studies have been conducted to elaborate on the impacts of traffic information dissemination on toll road utilization. In this study, 716 individuals completed a survey to measure representative public opinions and preferences for toll road usage in support of various traffic information dissemination classified by different modes, contents, and timeliness categories. A nested logit model was developed and estimated to identify the significant attributes of traffic information dissemination, traveler commuting patterns, routing behavior, and demographic characteristics, and analyze their impacts on toll road utilization. The results revealed that the travelers using dynamic message sign systems as their primary mode of receiving traffic information are more likely to choose toll roads. The potential toll road users also indicated their desire to obtain traffic information via internet. Information regarding accident locations, road hazard warnings, and congested roads is frequently sought by travelers. Furthermore, high-quality congested road information dissemination can significantly enhance travelers’ preferences of toll road usage. Specifically the study found that travelers anticipated an average travel time saving of about 11.3 min from better information; this is about 30 % of travelers’ average one-way commuting time. The mean value of the time savings was found to be about $11.82 per hour, close to ½ of the average Austin wage rate. The model specifications and result analyses provide in-depth insights in interpreting travelers’ behavioral tendencies of toll road utilization in support of traffic information. The results are also helpful to shape and develop future transportation toll system and transportation policy.     

A behavioral choice model of the use of car-sharing and ride-sourcing services

There are a number of disruptive mobility services that are increasingly finding their way into the marketplace. Two key examples of such services are car-sharing services and ride-sourcing services. In an effort to better understand the influence of various exogenous socio-economic and demographic variables on the frequency of use of ride-sourcing and car-sharing services, this paper presents a bivariate ordered probit model estimated on a survey data set derived from the 2014–2015 Puget Sound Regional Travel Study. Model estimation results show that users of these services tend to be young, well-educated, higher-income, working individuals residing in higher-density areas. There are significant interaction effects reflecting the influence of children and the built environment on disruptive mobility service usage. The model developed in this paper provides key insights into factors affecting market penetration of these services, and can be integrated in larger travel forecasting model systems to better predict the adoption and use of mobility-on-demand services.     

Estimation of multinomial probit-kernel integrated choice and latent variable model: comparison on one sequential and two simultaneous approaches

Integrated choice and latent variable (ICLV) model incorporates latent factors into standard discrete choice model with aim to provide greater explanatory power. Using simulated datasets, this study makes a comparison among three estimation approaches corresponding to the sequential approach and two simultaneous approaches including the maximum simulated likelihood with GHK estimator and maximum approximate composite marginal likelihood (MACML) approach, to evaluate their abilities to recover the underlying parameters of multinomial probit-kernel ICLV model. The results show that both simultaneous approaches outperform the sequential approach in terms of estimates accuracy and efficiency irrespective of the sample sizes, and the MACML approach is the most preferable due to its best performance on recovering true values of parameters with relatively small standard errors, especially when the sample size is large enough.