Further evidence on the temporal stability of disaggregate travel demand models

Current evidence on the transferability of disaggregate travel demand models is inconclusive. Adding to this body of research, the present analysis focuses upon the temporal characteristics of work trip behavior in the San Francisco Bay Area. Using before and after data sets associated with the BART Impact Travel Study, multinomial logit models of work trip modal choice are estimated. The results indicate that the general form and the coefficient estimates of a pre BART model are transferable in time. Moreover, when updated to reflect BART's presence, the model's predictive success and its implied elasticity measures are generally accurate, relative to those implied by reestimating the entire model on post BART data. Finally, as economic theory would predict, elasticity measures of the service related variables were found to increase over time.     

Temporal transferability of updated alternative-specific constants in disaggregate mode choice models

Methods of updating disaggregate discrete choice models have been proposed as a means of obtaining better transferability. However, the temporal transferability of models updated for better spatial transferability has rarely been analysed, and the factors affecting temporal transferability have not been determined. This paper deals with one updating method—the use of disaggregate data to update alternative-specific constants—and investigates the factors affecting the temporal transferability of the updated constants. In the analysis, repeated cross-section data collected in the Chukyo metropolitan area are divided, efficiently generating many application areas. The analysis showed that the factors can depend on regional characteristics and past travel behaviours (inertia), and are anti-symmetric and path-dependent of changes in the level of service.     

Price-based regulation of oligopolistic markets under discrete choice models of demand

We propose a framework to find optimal price-based policies to regulate markets characterized by oligopolistic competition and in which consumers make a discrete choice among a finite set of alternatives. The framework accommodates general discrete choice models available in the literature in order to capture heterogeneous consumer behavior. In our work, consumers are utility maximizers and are modeled according to random utility theory. Suppliers are modeled as profit maximizers, according to the traditional microeconomic treatment. Market competition is modeled as a non-cooperative game, for which an approximate equilibrium solution is sought. Finally, the regulator can affect the behavior of all other agents by giving subsidies or imposing taxes to consumers. In transport markets, economic instruments might target specific alternatives, to reduce externalities such as congestion or emissions, or specific segments of the population, to achieve social welfare objectives. In public policy, different agents have different individual or social objectives, possibly conflicting, which must be taken into account within a social welfare function. We present a mixed integer optimization model to find optimal policies subject to supplier profit maximization and consumer utility maximization constraints. Then, we propose a model-based heuristic approach based on the fixed-point iteration algorithm that finds an approximate equilibrium solution for the market. Numerical experiments on an intercity travel case study show how the regulator can optimize its decisions under different scenarios.

     

Testing probabilistic discrete choice models of travel demand by comparing predicted and observed aggregate choice shares

Probabilistic discrete choice models of travel demand often are tested for the presence of specification errors by comparing the models' predictions of aggregate choice shares in population strata with observed shares. A model is rejected as misspecified if the differences between its predictions and the observations are judged too large. This judgement usually is made on intuitive grounds without use of formal statistical methods and, therefore includes no systematic method for distinguishing the effects of specification errors on differences between predictions and observations from those of random sampling errors. This paper represents formal statistical tests for comparing predicted and observed aggregate chioce shares in population strata and reports the results of an investigation of the power of the tests. The test statistics are asymptotically χ² disturbed when the model being tested is correctly specified. The results of the power investigation suggests that greater power is obtained (i.e. there is ability to detect misspecified models) when all of the available data are used for both parameter estimation and specification testing than when the available data are divided into separate estimation and test data sets. Specification tests based on comparisons of predicted and observed aggregate choice shares appear to have less power than do likelihood ratio and likelihood ratio index specification tests when the alternative models required by the latter tests are correctly or approximately correctly specified. However, tests based on comparisons of predicted and observed shares ca have greater power than the other tests when the alternative models are seriouslymisspecified.     

Recent directions in automobile demand modeling

We identify seven areas in which recent research on auto demand has made significant progress. We discuss the importance of these areas in understanding and predicting auto demand and describe the methods that researchers have used to address them. Finally, we identify the issues that this research indicates require further investigation.     

Noisy demand and mode choice

Mode choice under stochastically varying demand is studied via a dynamic mathematical model which describes the behavioural interactions between population groups. The model is developed by assuming competing attractivity functions for automobile and public transit which motivate their use subject to an overall demand for transportation. When this demand is allowed to vary stochastically, a set of stochastic differential equations describing the model are obtained. These are solved for their steady-state values. It is found that noisy demand can structure the system qualitatively differently than when the demand is fixed. The noise is found to generally reduce the level of public transit ridership, but it also changes the values of the threshold at which new regimes occur and, most interestingly, it induces new steady-state solutions for ridership at critical values of the variance of demand. In the latter case, noise becomes a source of new possibilities in the system by triggering a steady-state solution not present in the noise-free environment.     

Forecasting light-duty vehicle demand using alternative-specific constants for endogeneity correction versus calibration

We investigate parameter recovery and forecast accuracy implications of incorporating alternative-specific constants (ASCs) in the utility functions of vehicle choice models. We compare two methods of incorporating ASCs: (1) a maximum likelihood estimator that computes ASCs post-hoc as calibration constants (MLE-C) and (2) a generalized method of moments estimator that uses instrumental variables (GMM-IV) to correct for price endogeneity. In a synthetic study we observe significant coefficient bias with MLE-C when the price-ASC correlation (endogeneity) is large. GMM-IV successfully mitigates this bias given valid instruments but exacerbates the bias given invalid instruments. Despite greater coefficient bias, MLE-C yields better forecasts than GMM-IV with valid instruments in most of the cases examined, including most cases where the price-ASC correlation present in the estimation data is absent in the prediction data. In a market study of U.S. midsize sedan sales from 2002 – 2006 the GMM-IV model predicts the 1-year-forward market better, but the MLE-C model predicts the 5-year-forward market better. Including an ASC in predictions by any of the methods proposed improves share forecasts, and assuming that the ASC of each new vehicle matches that of its closest competitor vehicle yields the best long term forecasts. We find evidence that the instruments most frequently used in the automotive demand literature may be invalid.     

Estimating the demand for electric automobiles using fully disaggregated probabilistic choice analysis

This study uses probabilistic choice models to predict potential demand for electric cars. Survey data are employed to estimate separate utility functions for each of 51 subjects. This provides a sample distribution of consumer preferences for vehicle attributes including price, operating cost and range. The results indicate great diversity in individual trade-offs among attributes, with range and top speed generally being highly valued. The sample of utility functions is then used to predict potential market shares for various kinds of electric vehicles as second cars. Demand is quite limited, except when (a) electric cars are considerably more advanced than anything likely to be available in the near future, and (b) consumers fear massive gasoline shortages. The latter effect derives from an observed “bias” in favor of electric autos, which is plausibly interpreted as a hedge against disruptions in the gasoline market.     

Forecasting passenger travel demand — international aspects

The differing demand forecasting approaches adopted on a variety of European international travel studies are described in this paper. These include four models used on the studies of the Paris-Brussels-Cologne/Amsterdam high speed rail line, the model used in appraising the proposed Channel Tunnel rail services and the EC TASC model system. Contrasting features of these forecasting procedures are highlighted.     

Estimating spatial interdependence in automobile type choice with survey data

In this article, we show that vehicle type ownership is spatially dependent at both the regional and household-level even after controlling for income and population density. We discuss reasons for the existence of spatial effects in vehicle ownership, and note potential implications for policymakers. Our results point to the importance of spatial relationships in transportation research and highlight the hazards of ignoring their role in affecting transportation outcomes. For example, if vehicle type choice is affected by neighborhood spillovers, agencies that regulate traffic flow and road safety could tailor their choice projections and policy tools to account for such interdependence.     

A framework for analyzing rank-ordered data with application to automobile demand

In this paper, we develop a general random utility framework for analyzing data on individuals’ rank-orderings. Specifically, we show that in the case with three alternatives one can express the probability of a particular rank-ordering as a simple function of first choice probabilities. This framework is applied to specify and estimate models of household demand for conventional gasoline cars and alternative fuel vehicles in Shanghai based on rank-ordered data obtained from a stated preference survey. Subsequently, the framework is extended to allow for random effects in the utility specification to allow for intrapersonal correlation in tastes across stated preference questions. The preferred model is then used to calculate demand probabilities and elasticities and the distribution of willingness-to-pay for alternative fuel vehicles.     

Estimating choice models containing attraction variables

The paper presents a family of disaggregate choice models, which are shown to be equivalent to many of the aggregate models commonly used in planning studies. A brief summary is given of the method that has been developed for estimating the parameters of these models. A generalisation is then introduced in which variables representing the attractiveness in terms of size or quantity of each alternative are allowed to enter the model. It is shown that the form in which these variables enter the model requires a more general estimation algorithm than is commonly used, and such an algorithm is presented. A series of practical tests of the new algorithm is described.     

An analysis of destination choice for opaque airline products using multidimensional binary logit models

We investigate how customers respond to an opaque airline product offered by a European carrier. In this opaque product design, customers are randomly assigned to travel to one of approximately ten destinations; however, for a fee they may exclude one or more destinations from the choice set (or a particular package design) prior to learning which destination they will travel to. We use a multidimensional binary logit model to predict the probability that one or more alternatives will be chosen by a customer. Results show that customers are more likely to pay to exclude destinations located close to the origin airport and destinations that speak the same language as the origin airport. Length of stay, cost of living at the destination, and measures of destination attractiveness are also found to be significant predictors for some package designs. Based on these findings, we offer general recommendations for how to design opaque packages for airline customers.     

Defining domains for models of travel demand

Travel demand models implicitly assume that people respond to changes in a continuous way. This is in contrast to the physical sciences, where discontinuous response is a common phenomenon and is embodied in such concepts as sub-critical and supercritical states.Recent studies have shown that responses to transport policies differ in degree and kind according to the nature and severity of the stimulus and the types of people affected. Response patterns may be categorised by the extent to which they involve adjustments to spatio-temporal or inter-personal linkages. This paper identifies four response domains, with a further distinction between permissive and forced changes.Most travel demand models are designed to operate within an independent, forced (and to a less extent independent permissive) domain and their forecasts become unreliable when responses lie outside that domain. Conversely, a model designed for a more complex domain is unnecessarily cumbersome where simpler responses apply. This paper describes the types of model which are appropriate for each domain and discusses how the effects of a policy may be assigned to the correct domain(s).     

Goodness-of-fit measures for probabilistic travel demand models

This paper is concerned with the assessment of the goodness-of-fit of nonlinear models of the type currently being used in the development of the disaggregate, behavioral travel demand approach. These models are emerging as a potential new technique for many transportation planning problems, although much research is yet needed before they are sufficiently developed for operational use. In order to pursue the necessary research, and also for the later assessment of operational models, it is necessary to have adequate measures of the goodness-of-fit.The paper examines the adequacy of standard measures of goodness-of-fit as applied to any nonlinear estimating equation and they are found to be inappropriate and inadequate. A little-known statistic, called the correlation ratio, is then defined and derived, and is explored as a substitute for the standard measures. In both theoretical and empirical tests, the correlation ratio is found to be a significantly more useful and appropriate measure of goodness-of-fit.Some further properties of the correlation ratio are examined, and the ratio is found to possess some degree of arbitrariness when applied to typical travel demand models. This arbitrariness, however, only impairs the usefulness of the correlation ratio in the absolute assessment of a model, but not for the comparative assessment of two or more models. Finally, a number of research tasks, relating to the correlation ratio, are identified.     

The analysis of travel and emission impacts of travel demand management strategies using activity-based models

This paper demonstrates, tests and shows the value of activity-based travel demand models and household sample enumeration forecasting techniques in evaluating the transportation and air quality impacts of travel demand management strategies. Using data from the Portland, Oregon metropolitan area, three transportation policies were evaluated both individually and in combination: transit improvements, pricing, and telecommunications. The activity-based models used in this testing represents a significant improvement to today's "four-step" sequential model systems by providing a deeper insight into the individual decision making process in response to transportation policies. A wider range of impacts is predicted, and indirect effects as well as synergistic effects of such policies are taken into consideration. These models are capable of providing the information needed to improve the linkage of transportation models with emissions and air quality analysis methodologies by improving the prediction of variables that are important to accurately estimating emissions and air quality impacts of transportation actions.     

Car fuel-type choice under travel demand management and economic incentives

Future levels of vehicle air pollution in urban areas will depend on the proportion of new car buyers who opt for less polluting vehicles, as these appear on the market. This paper examines the factors likely to influence the demand for lower emission and zero emission vehicles. Using a discrete choice experiment, suburban driver commuters choose between three types of vehicle, one conventional, one fuel-efficient and one electric. Each is characterized by varying vehicle cost and performance measures, range and refueling rates, and commuting costs and times. The latter are manipulated to determine how their use as economic instruments might influence vehicle choice. All cost and time variables are expressed as ratios of the respondent’s current situation. Parameters of a multinomial discrete choice model are used in a choice simulator to estimate the average choice probability of each type of vehicle under different scenarios reflecting possible future relative vehicle prices and performance levels as well as differential commuting costs and times based on policies aimed at encouraging the purchase of cleaner vehicles. The evidence is that the latter economic instruments will have modest effects on vehicle choice. By contrast there would be a large shift of demand to cleaner and zero-emission vehicles provided their cost and performance came within an acceptable range of conventional vehicles.     

Investigation into factors affecting the route choice in “Rijnstreek-West” with the aid of a disaggregate logit model

Transportation - Based on an analysis of observed automobile routes, an investigation is made into the factors affecting choice of routes. As in previous studies, the attempt to determine the...     

Incorporating social interaction into hybrid choice models

The aim of this paper is to develop a methodological framework for the incorporation of social interaction effects into choice models. The developed method provides insights for modeling the effect of social interaction on the formation of psychological factors (latent variables) and on the decision-making process. The assumption is based on the fact that the way the decision maker anticipates and processes the information regarding the behavior and the choices exhibited in her/his social environment, affects her/his attitudes and perceptions, which in turn affect her/his choices. The proposed method integrates choice models with decision makers’ psychological factors and latent social interaction. The model structure is simultaneously estimated providing an improvement over sequential methods as it provides consistent and efficient estimates of the parameters. The methodology is tested within the context of a household aiming to identify the social interaction effects between teenagers and their parents regarding walking-loving behavior and then the effect of this on mode to school choice behavior. The sample consists of 9,714 participants aged from 12 to 18 years old, representing 21 % of the adolescent population of Cyprus. The findings from the case study indicate that if the teenagers anticipate that their parents are walking lovers, then this increases the probability of teenagers to be walking-lovers too and in turn to choose walking to school. Generally, the findings from the application result in: (a) improvements in the explanatory power of choice models, (b) latent variables that are statistically significant, and (c) a real-world behavioral representation that includes the social interaction effect.     

Random covariance heterogeneity in discrete choice models

In this paper, we extend the standard discrete choice modelling framework by allowing for random variations in the substitution patterns between alternatives across respondents, leading to increased model flexibility. The paper shows how such a Mixed Covariance model can be specified either with purely random variation or with a mixture of random and deterministic variation. Additionally, the model can be based on an underlying GEV or ECL structure. Finally, the model can be specified as a continuous mixture or as a discrete mixture. An application on Stated Preference data for the choice of departure time and travel mode shows that important gains in model performance can be obtained by allowing for random covariance heterogeneity. Furthermore, the approach leads to significant differences in the implied willingness to pay measures, and the substitution patterns between alternatives.