Discrete choice with spatial correlation: A spatial autoregressive binary probit model with endogenous weight matrix (SARBP-EWM)

Discrete choice modeling is widely applied in transportation studies. However, the need to consider correlation between observations creates a challenge. In spatial econometrics, a spatial lag term with a pre-defined weight matrix is often used to capture such a correlation. In most previous studies, the weight matrix is assumed to be exogenous. However, this assumption is invalid in many cases, leading to biased and inconsistent parameter estimates. Although some attempts have been made to address the endogenous weight matrix issue, none has focused on discrete choice modeling. This paper fills an existing gap by developing a Spatial Autoregressive Binary Probit Model with Endogenous Weight Matrix (SARBP-EWM). The SARBP-EWM model explicitly considers the endogeneity by using two equations whose error terms are correlated. Markov Chain Monte Carlo (MCMC) method is used to estimate the model. Model validation with simulated data shows that the model parameters can converge to their true values and the endogenous weight matrix can be reliably recovered. The model is then applied to a simplified firm relocation choice problem, assuming that similar size firms influence one another. The model quantifies the peer effect, and takes into consideration other independent variables including industry type and population density. The estimation results suggest that peer influence among firms indeed affect their relocation choices. The application results offer important insights into business location choice and can inform future policy making. The sample size for applying the model is currently limited to hundreds of observations. This paper contributes to the existing literature on discrete choice modeling and spatial econometrics. It provides a new tool to discover spatial correlations that are hidden in a wide range of transportation issues, such as land development, location choice, and various travel behavior. Those hidden spatial correlations are otherwise difficult to identify and estimation results may be biased. Establishing a new model that explicitly considers endogenous weight matrix and applying the model to a real life transportation issue represent a significant contribution to the body of literature.     

Can you ever be certain? Reducing hypothetical bias in stated choice experiments via respondent reported choice certainty

Stated choice experiments are a preeminent method for researchers and practitioners who seek to examine the behavior of consumers. However, the extent to which these experiments can replicate real markets continues to be debated in the literature, with particular reference to the potential for biased estimates as a result of the hypothetical nature of such experiments. In this paper, a first in the transportation literature, we compare stated choice responses to revealed preference behavior and examine three methods proposed in the literature for calibrating choice experiments via reported choice certainty. In doing so we provide evidence that the incorrect calibration of responses can produce stated choice results that are more biased than doing nothing at all, however we show that by jointly estimating choice and choice certainty there is a significant reduction in hypothetical bias such that stated choice responses more directly replicate real behavior.     

Recovery of inter- and intra-personal heterogeneity using mixed logit models

Most applications of discrete choice models in transportation now utilise a random coefficient specification, such as mixed logit, to represent taste heterogeneity. However, little is known about the ability of these models to capture the heterogeneity in finite samples (as opposed to asymptotically). Also, due to the computational intensity of the standard estimation procedures, several alternative, less demanding methods have been proposed, and yet the relative accuracy of these methods has not been investigated. This is especially true in the context of work looking at joint inter-respondent and intra-respondent variation. This paper presents an overview of the various different estimators, gives insights into some of the theoretical properties, and analyses their performance in a large scale study on simulated data. In particular, we specify 31 different forms of heterogeneity, with multiple versions of each dataset, and with results from over 16,000 mixed logit estimation runs. The findings suggest that variation in tastes over consumers is captured by all the methods, including the simpler versions, at least when sample size is sufficiently large. When tastes vary over choice situations for each consumer, as well as over consumers, the ability of the methods to capture and differentiate the two sources of heterogeneity becomes more tenuous. Only the most computationally intensive approach is able to capture adequately the two sources of variation, but at the cost of very high run times. Our results highlight the difficulty of retrieving taste heterogeneity with only cross-sectional data, providing further evidence of the benefits of repeated choice data. Our findings also suggest that the data requirements of random coefficients models may be more substantial than is commonly assumed, further reinforcing concerns about small sample issues.     

A probabilistic map matching method for smartphone GPS data

Smartphones have the capability of recording various kinds of data from built-in sensors such as GPS in a non-intrusive, systematic way. In transportation studies, such as route choice modeling, the discrete sequences of GPS data need to be associated with the transportation network to generate meaningful paths. The poor quality of GPS data collected from smartphones precludes the use of state of the art map matching methods. In this paper, we propose a probabilistic map matching approach. It generates a set of potential true paths, and associates a likelihood with each of them. Both spatial (GPS coordinates) and temporal information (speed and time) is used to calculate the likelihood of the data for a specific path. Applications and analyses on real trips illustrate the robustness and effectiveness of the proposed approach. Also, as an application example, a Path-Size Logit model is estimated based on a sample of real observations. The estimation results show the viability of applying the proposed method in a real route choice modeling context.     

Analyzing high speed rail passengers’ train choices based on new online booking data in China

This study explores two nonparametric machine learning methods, namely support vector regression (SVR) and artificial neural networks (ANN), for understanding and predicting high-speed rail (HSR) travelers’ choices of ticket purchase timings, train types, and travel classes, using ticket sales data. In the train choice literature, discrete choice analysis is the predominant approach and many variants of logit models have been developed. Alternatively, emerging travel choice studies adopt non-utility-based methods, especially nonparametric machine learning methods including SVR and ANN, because (1) those methods do not rely on assumptions on the relations between choices and explanatory variables or any prior knowledge of the underlying relations; (2) they have superb capabilities of iteratively identifying patterns and extracting rules from data. This paper thus contributes to the HSR train choice literature by applying and comparing SVR and ANN with a real-world case study of the Shanghai-Beijing HSR market in China. A new normalized metric capturing both the load factor and the booking lead time is proposed as the target variable and several train service attributes, such as day of week, departure time, travel time, fare, are identified as input variables. Computational results demonstrate that both SVR and ANN can predict the train choice behavior with high accuracy, outperforming the linear regression approach. Potential applications of this study, such as rail pricing reform, have also been identified.     

Strategic sampling for large choice sets in estimation and application

Many discrete choice contexts in transportation deal with large choice sets, including destination, route, and vehicle choices. Model estimation with large numbers of alternatives remains computationally expensive. In the context of the multinomial logit (MNL) model, limiting the number of alternatives in estimation by simple random sampling (SRS) yields consistent parameter estimates, but estimator efficiency suffers. In the context of more general models, such as the mixed MNL, limiting the number of alternatives via SRS yields biased parameter estimates. In this paper, a new, strategic sampling scheme is introduced, which draws alternatives in proportion to updated choice-probability estimates. Since such probabilities are not known a priori, the first iteration uses SRS among all available alternatives. The sampling scheme is implemented here for a variety of simulated MNL and mixed-MNL data sets, with results suggesting that the new sampling scheme provides substantial efficiency benefits. Thanks to reductions in estimation error, parameter estimates are more accurate, on average. Moreover, in the mixed MNL case, where SRS produces biased estimates (due to violation of the independence of irrelevant alternatives property), the new sampling scheme appears to effectively eliminate such biases. Finally, it appears that only a single iteration of the new strategy (following the initialization step using SRS) is needed to deliver the strategy’s maximum efficiency gains.     

Analysis of hurricane evacuee mode choice behavior

The purpose of this study is to explain the evacuee mode choice behavior of Miami Beach residents using survey data from a hypothetical category four hurricane to reveal different evacuees’ plans. Evacuation logistics should incorporate the needs of transit users and car-less populations with special attention and proper treatment. A nested logit model has been developed to explain the mode choice decisions for evacuees’ from Miami Beach who use non-household transportation modes, such as special evacuation bus, taxi, regular bus, riding with someone from another household and another type of mode denoted and aggregated as other. Specifically, the model explains that the mode choice decisions of evacuees’, who are likely to use different non-household transportation modes, are influenced by several determining factors related to evacuees’ socio-demographics, household characteristics, evacuation destination and previous experience. The findings of this study will help emergency planners and policy-makers to develop better evacuation plans and strategies for evacuees depending on others for their evacuation transportation.     

The joint estimation of respondent-reported certainty and acceptability with choice

In the stated choice literature, increasing attention has been paid to methods that seek to close the gap between the choices from these experiments and the choices experienced in the real world. Attempts to produce model estimates that are truer to real market behaviours are especially important for transportation, where many important policy decisions rely on such experiments. A recent approach that has emerged makes use of a certainty index whereby respondents report how certain they are about each choice they make. Additional literature also posits that when making decisions, people first identify an acceptable set of alternatives (alternative acceptability) such that a consideration set if formed and it is from this reduced set that the ultimate choice is made. This paper presents two models that jointly estimates choice and choice certainty and choice and alternative acceptability. This joint estimation allows the modeller to overcome potential endogeneity that may exist between these responses. In comparing choices of differing certainty, surprisingly little difference in marginal sensitivities are found. This is not the case in the alternative acceptability models however. An important finding of this research is that what could be interpreted as preference heterogeneity may in fact be more closely linked to scale. The ramifications of these results on future research are discussed.     

Fleeing from Hurricane Irma: Empirical Analysis of Evacuation Behavior Using Discrete Choice Theory

This paper analyzes the observed decision-making behavior of a sample of individuals impacted by Hurricane Irma in 2017 (n = 645) by applying advanced methods based in discrete choice theory. Our first contribution is identifying population segments with distinct behavior by constructing a latent class choice model for the choice whether to evacuate or not. We find two latent segments distinguished by demographics and risk perception that tend to be either evacuation-keen or evacuation-reluctant and respond differently to mandatory evacuation orders.Evacuees subsequently face a multi-dimensional choice composed of concurrent decisions of their departure day, departure time of day, destination, shelter type, transportation mode, and route. While these concurrent decisions are often analyzed in isolation, our second contribution is the development of a portfolio choice model (PCM), which captures decision-dimensional dependency (if present) without requiring choices to be correlated or sequential. A PCM reframes the choice set as a bundle of concurrent decision dimensions, allowing for flexible and simple parameter estimation. Estimated models reveal subtle yet intuitive relations, creating new policy implications based on dimensional variables, secondary interactions, demographics, and risk-perception variables. For example, we find joint preferences for early-nighttime evacuations (i.e., evacuations more than three days before landfall and between 6:00 pm and 5:59 am) and early-highway evacuations (i.e., evacuations more than three days before landfall and on a route composed of at least 50% highways). These results indicate that transportation agencies should have the capabilities and resources to manage significant nighttime traffic along highways well before hurricane landfall.     

Big data in public transportation: a review of sources and methods

ABSTRACT

The collection of big data, as an alternative to traditional resource-intensive manual data collection approaches, has become significantly more feasible over the past decade. The availability of such data, coupled with more sophisticated predictive statistical techniques, has contributed to an increase in attention towards the application of these data, particularly for transportation analysis. Within the transportation literature, there is a growing emphasis on developing sources of commonly collected public transportation data into more powerful analytical tools. A commonly held belief is that application of big data to transportation problems will yield new insights previously unattainable through traditional transportation data sets. However, there exist many ambiguities related to what constitutes big data, the ethical implications of big data collection and application, and how to best utilize the emerging data sets. The existing literature exploring big data provides no clear and consistent definition. While the collection of big data has grown and its application in both research and practice continues to expand, there is a significant disparity between methods of analysis applied to such data. This paper summarizes the recent literature on sources of big data and commonly applied methods used in its application to public transportation problems. We assess predominant big data sources, most frequently studied topics, and methodologies employed. The literature suggests smart card and automated data are the two big data sources most frequently used by researchers to conduct public transit analyses. The studies reviewed indicate that big data has largely been used to understand transit users’ travel behavior and to assess public transit service quality. The techniques reported in the literature largely mirror those used with smaller data sets. The application of more advanced statistical methods, commonly associated with big data, has been limited to a small number of studies. In order to fully capture the value of big data, new approaches to analysis will be necessary.     

Dynamic process model of mass effects on travel demand

Whereas transportation planners commonly predict the negative impacts of mass transportation, there is increasing empirical evidence of the existence of positive mass effects, whereby increased use of a mode by the ‘mass’ will generally increase its attractiveness for future travellers. In this paper we consider the dynamic impact of such an effect on the problem of travel demand forecasting, with particular regards to social network effects. Our proposed modelling approach is inspired by literature from social physics, evolutionary game theory and marketing. For simplicity of exposition, our model is specified for a scenario in which (a) there is a binary choice between two mobility lifestyles, referred to as car-oriented and transit-oriented, and (b) there are two population groups, where one is the “leading” or “innovative” population group and the other the “following” or “imitating” population group. This latter distinction follows the rather well-known Bass model from the marketing literature (1969). We develop the transition probabilities and transition dynamics. We illustrate with a numerical case study that despite lower intrinsic utility for the transit lifestyle, significant changes towards this lifestyle can be achieved by considering congestion, service improvements and mass effects. We further illustrate that mass effects can be positive or negative. In all cases we explore the sensitivity of our conclusions to the assumed parameter values.     

The effects of attitudes and personality traits on mode choice

We hypothesise that differences in people’s attitudes and personality traits lead them to attribute varying importance to environmental considerations, safety, comfort, convenience and flexibility. Differences in personality traits can be revealed not only in the individuals’ choice of transport, but also in other actions of their everyday lives—such as how much they recycle, whether they take precautions or avoid dangerous pursuits. Conditioning on a set of exogenous individual characteristics, we use indicators of attitudes and personality traits to form latent variables for inclusion in an, otherwise standard, discrete mode choice model. With a sample of Swedish commuters, we find that both attitudes towards flexibility and comfort, as well as being pro-environmentally inclined, influence the individual’s choice of mode. Although modal time and cost still are important, it follows that there are other ways, apart from economic incentives, to attract individuals to the, from society’s perspective, desirable public modes of transport. Our results should provide useful information to policy-makers and transportation planners developing sustainable transportation systems.     

Determining structural route components from GPS traces

Analysis of GPS traces shows that people often do not use the least cost path through the transportation network while making trips. This leads to the question which structural path characteristics can be used to construct realistic route choice sets for use in traffic simulation models. In this paper, we investigate the hypothesis that, for utilitarian trips, the route between origin and destination consists of a small number of concatenated least cost paths. The hypothesis is verified by analyzing routes extracted from large sets of recorded GPS traces which constitute revealed preference information. Trips have been extracted from the traces and for each trip the path in the transportation network is determined by map matching. This is followed by a path decomposition phase for which the algorithm constitutes the first contribution of this paper. There are multiple ways to split a given path in a directed graph into a minimal number of subpaths of minimal cost. By calculating two specific path splittings, it is possible to identify subsets of the vertices (splitVertexSuites) that can be used to generate every possible minimum path splitting by taking one vertex from each such subset. As a second contribution, we show how the extracted information is used in microscopic travel simulation. The distribution for the size of the minimum decomposition, extracted from the GPS traces, can be used in constrained enumeration methods for route choice set generation. The sets of vertices that can act as boundary vertices separating consecutive route parts contain way points (landmarks) having a particular meaning to their user. The paper explains the theoretical aspects of route splitting as well as the process to extract splitVertexSuites from big data. It reports statistical distributions extracted from sets of GPS traces for both multimodal person movements and unimodal car trips.     

Specification of regret-based models of choice behaviour: formal analyses and experimental design based evidence

Travel behaviour analysis has recently witnessed a rapidly growing interest in regret-based models of choice behaviour. Two different model specifications have been introduced in the transportation literature. Chorus et al. (Transportation Research B 42: 1–18, 2008a; in: Proceedings 87th Annual Meeting of the Transportation Research Board, Washington DC, 2008b) specified regret as a (non) linear function of the difference between the best-foregone choice alternative and the chosen alternative. Later, as an approximation to the original specification, Chorus (2010) suggested a logarithm function of utility differences between all choice alternatives, mainly for ease of estimation. This paper makes two contributions to this literature. First, formal analyses are conducted to identify the parameter space where the logarithmic specification becomes theoretically inferior to the original specification. Second, an empirical stated choice study on the choice of shopping centre is conducted to empirically test which specification best describes stated choices. Results suggest that for the collected data the original specification outperforms the new specification. Implications of this finding for the application of regret-based choice models in travel behaviour analysis are discussed.     

Misclassification in travel surveys and implications to choice modeling: application to household auto ownership decisions

Travel surveys that elicit responses to questions regarding daily activity and travel choices form the basis for most of the transportation planning and policy analysis. The response variables collected in these surveys are prone to errors leading to mismeasurement or misclassification. Standard modeling methods that ignore these errors while modeling travel choices can lead to biased parameter estimates. In this study, methods available in the econometrics literature were used to quantify and assess the impact of misclassification errors in auto ownership choice data. The results uncovered significant misclassification rates ranging from 1 to 40% for different auto ownership alternatives. Also, the results from latent class models provide evidence for variation in misclassification probabilities across different population segments. Models that ignore misclassification were not only found to have lower statistical fit but also significantly different elasticity effects for choice alternatives with high misclassification probabilities. The methods developed in this study can be extended to analyze misclassification in several response variables (e.g., mode choice, activity purpose, trip/tour frequency, and mileage) that constitute the core of advanced travel demand models including tour and activity-based models.

     

An integrated model for discrete and continuous decisions with application to vehicle ownership,type and usage choices

This paper proposes an integrated econometric framework for discrete and continuous choice dimensions. The model system is applied to the problem of household vehicle ownership, type and usage. A multinomial probit is used to estimate household vehicle ownership, a multinomial logit is used to estimate the vehicle type (class and vintage) choices, and a regression is used to estimate the vehicle usage decisions. Correlation between the discrete (number of vehicles) and the continuous (total annual miles traveled) parts is captured with a full variance–covariance matrix of the unobserved factors. The model system is estimated using Simulated Log-Likelihood methods on data extracted from the 2009 US National Household Travel Survey and a secondary dataset on vehicle characteristics. Model estimates are applied to evaluate changes in vehicle holding and miles driven, in response to the evolution of social societies, living environment and transportation policies.     

Can scale and coefficient heterogeneity be separated in random coefficients models?

There is growing interest in the notion that a significant component of the heterogeneity retrieved in random coefficients models may actually relate to variations in absolute sensitivities, a phenomenon referred to as scale heterogeneity. As a result, a number of authors have tried to explicitly model such scale heterogeneity, which is shared across coefficients, and separate it from heterogeneity in individual coefficients. This direction of work has in part motivated the development of specialised modelling tools such as the G-MNL model. While not disagreeing with the notion that scale heterogeneity across respondents exists, this paper argues that attempts in the literature to disentangle scale heterogeneity from heterogeneity in individual coefficients in discrete choice models are misguided. In particular, we show how the various model specifications can in fact simply be seen as different parameterisations, and that any gains in fit obtained in random scale models are the result of using more flexible distributions, rather than an ability to capture scale heterogeneity. We illustrate our arguments through an empirical example and show how the conclusions from past work are based on misinterpretations of model results.     

Survey and empirical evaluation of nonhomogeneous arrival process models with taxi data

Arrival processes are important inputs to many transportation system functions, such as vehicle prepositioning, taxi dispatch, bus holding strategies, and dynamic pricing. We conduct a comprehensive survey of the literature which shows that many transport systems employ basic homogeneous arrival process models or static nonhomogeneous processes. We conduct an empirical experiment to compare five state of the art arrival process short term prediction models using a common transportation system data set: New York taxi passenger pickups in 2013. Pickup data is split between 672 observations for model estimation and 96 observations for validation. From our experiment, we obtain evidence to support a recent model called FM‐IntGARCH, which is able to combine the benefits of both time series models and discrete count processes. Using a set of seven performance metrics from the literature, FM‐IntGARCH is shown to outperform the offline models—seasonal factor method, piecewise linear model—as well as the online models—ARIMA, Gaussian Cox process. Implications for operating data‐driven “smart” transit systems and urban informatics are discussed. Copyright © 2016 John Wiley & Sons, Ltd.     

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.     

Discrete/continuous econometric models and their application to transport analysis

A wide range of transport‐related decisions involve the linking of discrete choices (e.g. of vehicle choice) and continuous choices (e.g. of vehicle use). In recent years econometricians have developed procedures for integrating such choices into a framework that is both economically and statistically sound. The literature is however somewhat technical. The objective of this paper is to provide a general overview of the basic elements of discrete/continuous econometric modelling with an emphasis on transport applications. It is hoped that such an introduction will demonstrate that the essence of the approach for the practitioner is quite straightforward and can be implemented with widely available computer software.