Discrete Choice Approach Research Articles

Evaluation of discrete choice and neural network approaches for modelling driver compliance with traffic information

This article evaluates dynamic driver behaviour models that can be used, in the context of intelligent transport systems (ITS), to predict drivers’ compliance with traffic information. The inputs to this type of models comprise drivers’ individual socio-economic characteristics and other variables that may influence their compliance behaviour. The output is a binary integer representing whether drivers comply with travel advice or not. Two approaches are available for formulating this category of classification problems: discrete choice models and artificial neural networks (ANNs). The literature on this topic clearly points to the limitations of the discrete choice approach which suffers from assumptions of perfect information about travel conditions, infinite information processing capabilities of drivers and inability to model the uncertainty in driver decision making or the vagueness in information received from ITS devices. ANNs, on the other hand, are able to deal with complex non-linear relationships, are fault tolerant in producing acceptable results under imperfect inputs and are suitable for modelling reactive behaviour which is often described using rules, linking a perceived situation with appropriate action. This study aims to evaluate the performance of these two categories of models based on a common data set of driver behaviour, collected from a field behavioural survey on a congested commuting corridor in Brisbane, Australia. This article proposes the combination of fuzzy logic and neural networks as a viable approach for overcoming the limitations of existing algorithms by modelling drivers as heterogeneous individuals. The results showed superior performance for a neuro-fuzzy model over binary choice models in terms of classifying or predicting the categories of drivers most likely to comply (or not comply) with traffic advice. The accuracy of the proposed model, in terms of classification rate, ranged between 95 and 97% compared to 50–73% for the discrete choice models.

Formal Evaluation of the Impact of Barriers and Connectors on Residential Burglars' Macro-Level Offending Location Choices

Previous research evaluating burglars' offending location choices has produced mixed findings about the influence of physical barriers and connectors on offender movement patterns. Consequently, this article utilises the discrete spatial choice approach to formally evaluate the impact of barriers and connectors on residential burglars' macro-level offending location choices. Data from Perth, Western Australia, demonstrated that physical barriers and connectors exert significant influence on offender decision-making at this level, and that the influence of impermeable barriers increases with proximity of these obstacles to the offender's point of origin. These findings provide formalised evidence for the independent importance of physical barriers and connectors in offender movement and are discussed with respect to current environmental criminology theory.

Migration and hedonic valuation: The case of air quality

Conventional hedonic techniques for estimating the value of local amenities rely on the assumption that households move freely among locations. We show that when moving is costly, the variation in housing prices and wages across locations may no longer reflect the value of differences in local amenities. We develop an alternative discrete-choice approach that models the household location decision directly, and we apply it to the case of air quality in US metro areas in 1990 and 2000. Because air pollution is likely to be correlated with unobservable local characteristics such as economic activity, we instrument for air quality using the contribution of distant sources to local pollution—excluding emissions from local sources, which are most likely to be correlated with local conditions. Our model yields an estimated elasticity of willingness to pay with respect to air quality of 0.34–0.42. These estimates imply that the median household would pay $149–$185 (in constant 1982–1984 dollars) for a one-unit reduction in average ambient concentrations of particulate matter. These estimates are three times greater than the marginal willingness to pay estimated by a conventional hedonic model using the same data. Our results are robust to a range of covariates, instrumenting strategies, and functional form assumptions. The findings also confirm the importance of instrumenting for local air pollution.

On consumers’ willingness to purchase nutritionally enhanced genetically modified food

This study assesses consumers' willingness to purchase genetically modified (GM) food products with two different types of benefits: an input (i.e., reduced pesticides) and an output trait benefit (i.e., nutritionally enhanced). Data were collected using a telephone survey of an Italian households sample. Discrete choice approach is used to elicit the purchase intentions of the respondents. Four separate probit models are estimated to examine the effect of various factors on choices. Results suggest that majority of Italians are not willing to buy GM food products even if they are nutritionally enhanced. However, more consumers are willing to buy a nutritionally enhanced plant based GM product than a traditional plant based GM (with input trait benefit). Willingness to buy for a nutritionally enhanced animal based GM product and for a traditional animal based GM product with input trait benefit are similar. Consumers unwilling to buy GM food would not buy it even if it is nutritionally enhanced and has lower price. However, there is a niche of consumers who are willing to buy nutritionally enhanced GM food products even at a premium. Knowledge of science and trust in scientists consistently affect Italian consumers' willingness to buy the GM products.

Neighborhoods, cars, and commuting in New York City: A discrete choice approach

Cities around the world are trying out a multitude of transportation policy and investment alternatives with the aim of reducing car-induced externalities. However, without a solid understanding of how people make their transportation and residential location choices, it is hard to tell which of these policies and investments are really doing the job and which are wasting precious city resources. The focus of this paper is the determinants of car ownership and car use for commuting. Using survey data from 1997 to 1998 collected in New York City, this paper uses discrete choice econometrics to estimate a model of the choices of car ownership and commute mode while also modeling the related choice of residential location. The main story told by this analysis is that New Yorkers are more sensitive to changes in travel time than they are to changes in travel cost. The model predicts that the most effective ways to reduce both auto ownership and car commuting involve changing the relative travel times for cars and transit, making transit trips faster by increasing both the frequency and the speed of service and making auto trips slower – perhaps simply by allowing traffic congestion. Population density also appears to have a substantial effect on car ownership in New York.

An alternative interpretation of multiple bounded WTP data—Certainty dependent payment card intervals

This paper analyzes the multiple bounded format, in which uncertainty is directly incorporated into the WTP question. A new approach for analyzing multiple bounded uncertainty data is presented. The intuition underlying the approach is that uncertain individuals would like to state their WTP as intervals rather than precise values and that the width of the intervals is determined by the degree of uncertainty. The approach is compared to the one applied in Welsh and Poe [Welsh, M., Poe, G.L., 1998. Elicitation effects in contingent valuation: comparisons to a multiple bounded discrete choice approach. Journal of Environmental Economics and Management 36, 170–185] which treats uncertainty by conditioning responses on specific verbal probability statements. We argue that the conditioning approach overestimates mean and median WTP and that conditioning WTP estimates on probability statements like “probably” and “unsure” make them “fuzzy”. To empirically compare the two approaches we use data from 2004 concerning implementation of a predator protection policy in Sweden. Our analysis show that the suggested approach: (1) is more intuitive; (2) better fits the data; (3) estimates mean and median WTP with better precision; (4) is less sensitive to distributional assumptions; and (5) it is better suited for policy analysis.

Predicting customer choice in services using discrete choice analysis

This paper presents an overview of the science and art of discrete choice modeling for service sector applications. With the ongoing momentum of service science, management, and engineering, the discrete choice modeling approach provides a sophisticated tool kit for assessing the needs and preferences of service customers. We provide directions for designing and executing discrete choice studies for services and discuss several examples for a number of industries including health care, financial services, retail, hospitality, and online services. We conclude with a discussion of the many managerial implications of the discrete choice approach.

A Discrete-Choice Approach to Modeling Social Influence on Individual Decision Making

Individual decision making is commonly studied using discrete choice models. Models of this type are applied extensively to the study of travel behavior, residential location, and employment decisions, among other topics of interest. A notable characteristic of the underlying economic theory is the assumption that individuals seek to maximize utility on the basis of their personal attributes and the attributes of the alternatives available to them. This approach ignores the interrelated nature of decision making in social situations—in other words, the role that social structures play in shaping behavior. In this paper we describe a multinomial discrete choice approach to analyzing individual behavior in social situations where position in a social network may encourage or discourage different courses of action. By means of a simulation example, we explore some properties of the model, in particular the effect of network topology.

The Impact of an Urban Growth Boundary on Land Development in Knox County, Tennessee: A Comparison of Two-Stage Probit Least Squares and Multilayer Neural Network Models

The impact of an urban growth boundary (UGB) on land development in Knox County, TN is estimated via two-stage probit and neural-network models. The insignificance of UGB variable in the two-stage probit model and more visible development patterns in the western part of Knoxville and the neighboring town of Farragut during the post-UGB period in both models suggest that the UGB has not curtailed urban sprawl. Although the network model is found to be a viable alternative to more conventional discrete choice approach for improving the predictability of land development, it is at the cost of evaluating marginal effects.

Heterogeneity in Post-materialist Value Priorities. Evidence from a Latent Class Discrete Choice Approach

should be conceived of as relative priorities rather than absolute preferences. As such they insist on using ranking techniques of measurement which generates choice data. In this study, we aim at validating the measurement of Inglehart’s (post-)materialism by means of a latent class discrete choice model. We argue that from a statistical point of view this is the appropriate way of dealing with ranking data. Furthermore, the analyses revealed a heterogeneity in (post-)materialist value priorities that has previously been left unobserved. Consistent with Inglehart’s research a post-materialist class is discerned irrespective of the number of latent classes that is selected. However, as far as materialism is concerned three different types of materialist concerns can be distinguished. The validity of the empirical typology is further demonstrated by linking it to key covariates and political attitudes.

Does the Service Argument Justify Resale Price Maintenance?

The paper analyzes the welfare effects of resale price maintenance (RPM) when the level of service of one retailer generates negative rather than positive spillovers for other retailers. A simple linear demand system is used and is linked to preferences of consumers via a discrete-choice approach. Under this specification, and in contrast to the literature dealing with this case, RPM may increase service (and price) or decrease service (and price). More interestingly, whether service increases or decreases due to RPM provides no clue to whether welfare increases or decreases due to RPM. Either case may arise independently of the effect of RPM on service.

Alternative Approaches to Discrete Working Time Choice in an Age Framework

We compare two options of integrating discrete working time choice of heterogenous households into a general equilibrium model. The first, known from the literature, produces household heterogeneity through a working time preference parameter. We contrast this with a model that directly incorporates a logit discrete-choice approach into a AGE framework. On the grounds of both calibration consistency and adequate accomodation of within-household interaction, we argue that the logit approach is preferable.

Application of radial basis function and generalized regression neural networks in non-linear utility function specification for travel mode choice modelling

Application of soft computational methods, especially artificial neural networks, in examining individual traveller behaviour is not encountered frequently. In most of the relevant cited papers, the feed-forward back propagation neural network (FFBPNN) models or hybrid models of FFBPNNs are proposed. However the feed-forward back propagation algorithm has some drawbacks, which can easily lead the model to develop in an inaccurate direction. Throughout this study, two different algorithms, radial basis function neural network (RBFNN) and generalized regression neural network (GRNN), are employed to propose a new calibration process for travel mode choice analysis in a transportation modelling framework. The neural network methods are not applied directly to calibrate models but are used as a sub-process for alternative non-linear model specification on utility function. Results show both the surpassing of RBFNNs and GRNNs over frequently used FFBPNNs, and the superiority of neural network methods over a conventional statistical model, multivariate linear regression, during mode choice calibrations. Also having experienced the existence of a claim that ANNs can tackle the problem of travel choice modelling as well as, if not better than, the discrete choice approach [D.A. Hensher, T.T. Ton, A comparison of the predictive potential of artificial neural networks and nested logit models for commuter mode choice, Transp. Res., Part E Logist. Trans. Rev. 36 (3) (2000) 155–172], use of such soft computing tools in studying traveller behaviour should be an autonomous part of a calibration process.

How Do Residential Burglars Select Target Areas?

This paper introduces the discrete spatial choice approach to the study of criminal target choice. The approach is used to assess whether residential burglars are attracted to target areas that are affluent, accessible, and poorly guarded. In addition, the importance of these criteria is postulated to vary across burglars. The theory is tested using data on 548 residential burglaries, committed by 290 burglars from the city of The Hague, the Netherlands. The likelihood of a neighbourhood’s being selected for burglary is heightened by its ethnic heterogeneity, its percentage of single-family dwellings, and its proximity to where the offender lives. The results and prospects of the discrete spatial choice approach for spatial target selection research are discussed.

Dynamics of the federal funds target rate: a nonstationary discrete choice approach

AbstractWe apply a discrete choice approach to model the empirical behaviour of the Federal Reserve in changing the federal funds target rate, the benchmark of short‐term market interest rates in the US. Our methods allow the explanatory variables to be nonstationary as well as stationary. This feature is particularly useful in the present application as many economic fundamentals that are monitored by the Fed and are believed to affect decisions to adjust interest rate targets display some nonstationarity over time. The chosen model successfully predicts the majority of the target rate changes during the time period considered (1994–2001) and helps to explain strings of similar intervention decisions by the Fed. Based on the model‐implied optimal interest rate, our findings suggest that there is a lag in the Fed's reaction to economic shocks during this period. Copyright © 2004 John Wiley & Sons, Ltd.

Payment Certainty in Discrete Choice Contingent Valuation Responses: Results from a Field Validity Test

1. Introduction Markets do exist to provide the information necessary for conducting benefit-cost analyses in many public policy decision-making situations. When desired estimates of benefits or costs manifest in the market (Arrow 1999, p. vi), economists have increasingly turned to contingent valuation surveys to elicit the values that individuals would place on public goods and externalities (Mitchell and Carson 1989; Cropper and Oates 1992; Deacon et al. 1998). Although discrete choice, take-it-or-leave-it methods of eliciting preferences have gained favor on theoretical grounds (Arrow et al. 1993; Carson, Groves, and Machina 1999) and realism (Hanemann 1994), accumulated evidence from a number of laboratory and field contingent valuation validity studies suggests that these methods overstate actual willingness to pay (WTP) for private and public goods (e.g., Cummings, Harrison, and Rustrom 1995; Brown et al. 1996; Cummings et al. 1997; Balistreri et al. 2001; Champ and Bishop 2001). That is, respondents are more likely to say to commitments than actual commitments, reflecting hypothetical bias and the need for contingent valuation responses (Harrison 2002). Two recent papers offer possible methods for calibrating discrete-choice responses by considering payment certainty levels reported by respondents. In what we term the follow-up certainty question (FCQ) method, Champ et al. (1997) ask dichotomous respondents to indicate how certain they are, on a scale from 1 (very uncertain) to (very certain), that they would pay the stated dollar amount if the program were actually offered. Separate WTP functions are estimated for each certainty level. Welsh and Poe (1998) instead adopt a discrete choice (MBDC) approach that directly incorporates certainty levels through a two-dimensional decision matrix: One dimension specifies dollar amounts that individuals would be required to pay on implementation of the policy, and the second dimension allows individuals to express their level of voting certainty through definitely no, probably no, not sure, probably yes, and definitely yes response options. A multiple-bounded logit model is used to estimate separate WTP functions for each certainty level. In this paper, we use a field validity test of contributions to a green electricity pricing program to further explore these methods and address several validity issues. First, using actual sign-up data as a criterion, we derive optimal correction strategies for the two methods. Previous laboratory research on private goods suggests that dichotomous responses from those who are definitely sure (Blumenschein et al. 1998) or at least probably sure (Johannesson et al. 1999) closely predict actual purchase decisions. Johannesson, Liljas, and Johansson (1998) find that respondents who are absolutely sure of their decision provide a conservative estimate of real purchases. These laboratory results are replicated in public goods contingent valuation field validity research using FCQ methods, suggesting that models that only use responses with certainty values on a 1-to-10 scale of 7 and higher (Ethier et al. 2000), 8 and higher (Champ and Bishop 2001), or 10 (Champ et al. 1997) best predict actual contributions. We are the first to provide correction strategies for the MDC approach. Second, we examine if the experimental classroom results reported in Welsh and Poe can be replicated in the field. In that paper, the authors compare estimated logistic response distributions from dichotomous questions and MBDC not sure responses and find that they are statistically different. This suggests that respondents who are uncertain of their values will tend to yea-say when asked a single dichotomous question, a result that has been replicated elsewhere (e. …

Household Choice of Drinking–Water Source in the Philippines

The present paper analyzes household choice of drinking–water source for 769 rural households in metropolitan Cebu, Philippines. The study aims to analyze the effects of input prices, tastes and household size on the choice probabilities. For the empirical analysis, a discrete–choice approach is employed, consistent with utility maximization. The findings indicate that the time cost is an important determinant of household choice of drinking–water source and, surprisingly, that taste, proxied by income, has ambiguous effects on household choice. The present study is contrasted to an earlier study in which inconsistency with utility maximization cannot be ruled out.

Physician choice of medical specialty: do economic incentives matter?

In this paper, a probabilistic discrete choice approach is used to examine the influence of economic and non-economic factors on the choice of medical specialty by new physicians. A two level nested logit model is estimated that relaxes the independence from irrelevant alternatives assumption, and allows for a potentially more realistic pattern of substitution across specialty choices. The results from this specification are compared to those obtained from joint and conditional logit models. The findings, which are relatively robust across models, indicate that economic incentives play an important role in the specialty choice process, in particular, expected hours worked and medical school indebtedness. Physician tastes for specialties also appear to have an important influence.

A comparison of the predictive potential of artificial neural networks and nested logit models for commuter mode choice

Abstract Research in the field of artificial intelligence systems has been exploring the use of artificial neural networks (ANN) as a framework within which many traffic and transport problems can be studied. One appeal of ANN is their use of pattern association and error correction to represent a problem. This contrasts with the random utility maximisation rule in discrete choice modelling. ANN enables a full set of human perceptions about a particular problem to be represented by artificial networks of neurons. A claim of ANN is that it can tackle the problem of travel demand forecasting and modelling as well if not better than the discrete choice approach. The use of such tools in studying individual traveller behaviour thus opens up an opportunity to consider the extent to which there are representation frameworks which complement or replace discrete choice methods. This paper explores the merits of neural networks by comparing the predictive capability of ANN and nested logit models in the context of commuter mode choice.

The use of conjoint analysis to elicit community preferences in public health research: a case study of hospital services in South Australia

AimsTo demonstrate the use of conjoint analysis (CA) in public health research through a survey of the South Australian community about aspects of their public hospital services. MethodsA series of focus groups determined the most important attributes in choice of hospital services. These were built into a CA survey, using the discrete choice approach. The survey was posted to a representative sample of 700 South Australians. Theoretical validity, internal consistency and non-response bias were all investigated. ResultsSome 231 individuals returned the questionnaire. The attribute, ‘improvement in complication rates’ was positively associated with choice of hospital. Three attributes were found to be negatively associated with such choice: ‘waiting times for casualty’, ‘waiting times for elective surgery’ and, anomalously, ‘parking and transport facilities’. Travel time’ and the cost attribute, ‘Medicare levy’ were not statistically significant. Trade-offs between the significant attributes were estimated, as were satisfaction or utility scores for different ways of providing hospital services. Results concerning internal consistency and internal validity were encouraging, but some potential for non-response bias was detected. ConclusionA high premium is placed on the quality of hospital care and members of the community are prepared to choose between hospitals largely on the basis of outcomes and length of waiting times for elective surgery and in casualty. ImplicationsCA can yield potentially policy-relevant information about community preferences for health services.