Random Choice Model Research Articles

Extending the Logit-Mixed Logit Model for a Combination of Random and Fixed Parameters

A semi-parametric logit-mixed logit (LML) model to specify the mixing distribution of the preference heterogeneity is the recent advancement in logit-type models. LML provides a general specification for many previous semi-parametric and non-parametric models. Additionally, a special form of the likelihood gradient makes LML computationally more efficient than other parametric models. However, the original LML formulation assumes all utility parameters to be random. This study extends LML to a combination of fixed and random parameters (LML-FR) and motivates such combination in random parameter choice models in general. We further show that the likelihood of LML-FR specification looses its special properties, leading to a much higher estimation time than that of the original LML specification. In an empirical application about preferences for alternative fuel vehicles in China, estimation time increased by a factor of 15-20 when introducing fixed parameters to the LML model. The comparison of the original LML and LML-FR specifications with other parametric models indicate that LML-FR model outperforms these parametric models in terms of model-fit-criteria (e.g., AIC, BIC, and likelihood), but the same does not hold for the original LML, i.e. when all parameters are assumed random.

Incorporating psycho-physical mapping into random regret choice models: model specifications and empirical performance assessments

Recently, regret-based choice models have been introduced in the travel behavior research community as an alternative to expected/random utility models. The fundamental proposition of regret theory is that individuals minimize the amount of regret they (are expected to) experience when choosing among choice alternatives. In this context, regret is defined as a function of attribute differences between the considered choice alternative and one or more foregone choice alternatives in an individual’s choice set. This definition of regret as a function of physical attributes of the choice alternatives implies that current regret-based choice models do not account for the perception of attributes, which is likely related to their magnitude. Therefore, in this paper, we propose and empirically test such an elaboration of the basic regret-minimization models. The current paper sets out to formulate random regret minimization models that incorporate a non-linear representation of the perception of attribute levels. Inspired by long-standing research on psycho-physical measurement, it is assumed that the perception of stimuli (attributes) is proportional to their magnitude. To allow for slight deviations from this representation, a more general non-linear psycho-physical representation of the relationship between attribute levels and their perception is also tested. The suggested models are tested using two data sets, one data set concerned the stated choice of shopping centre, the other concerned revealed preference of mode choice. The two newly formulated regret models are compared against each other and against their original random regret minimization base specifications. In addition to comparing the predictive performance of these model specifications, validation tests are conducted. In each case study, the newly suggested regret models, incorporating a non-linear representation of perception, achieve significant improvements in goodness-of-fit over the original regret formulations. The results of the K-fold validation tests provide further support to this finding.

Identifying Latent Classes of Pedestrian Crowd Evacuees

The tactical decision making of humans when they exit crowded built environments was investigated. Recent studies have shown that the relative weights and priorities that pedestrian evacuees associate with different factors contributing to their choice of exit vary significantly from person to person, yet this heterogeneity is not explainable by the demographics of the decision makers, such as gender. Hence, it is assumed that the choice could be attributable to the presence of latent classes (latent types of behavior) unassociated with observable demographic characteristics. The standard latent class and the random coefficient latent class choice models were implemented with data on stated exit choices to explore this. The latter model accommodated coefficient heterogeneity within each class by allowing the coefficients associated with each class to vary randomly. Data analyses showed that exit choice heterogeneity did not necessarily need to be represented by continuous distributions and could have been accommodated in a finite number of discrete categories. In fact, only two classes of evacuees (congestion avoiders and distance avoiders) could be identified. The terminology derived from the fact that the chief difference in evacuation behavior of the two classes was associated with their perception of crowding-related factors (i.e., exit density and directional density). The modeling results suggested that congestion avoiders outnumbered distance avoiders. The analyses also showed that the within-class heterogeneity faded into insignificance once the binary discrete classification was imposed: evacuees’ decision-making heterogeneity was sufficiently reducible into two discrete categories with negligible inner-class variation. This finding provides a more tangible interpretation of exit choice heterogeneity than the continuous mixture approach. The finding also simplifies the way in which the modeler deals with decision heterogeneity when the exit choice model is implemented for prediction purposes.

Estimation and application of a multi-class multi-criteria mixed paired combinatorial logit model for transport networks analysis

Probabilistic approach of transport network modelling has received significant attention in recent years. Despite the recent progress in this area, the full advantage of the potential capability of random utility choice models has not yet been fully realised. This research is intended to introduce a new approach of combining the state-of-the-art paired combinatorial logit route choice modelling and random coefficient choice models in traffic assignment. While the former addresses the problem of correlation among path utilities, the latter can capture the random taste heterogeneity in route choice decision-making. Including an additional monetary cost explanatory variable, the model would be able to assess a broader range of planning policies such as road pricing. In addition, distinguishing multiple classes of decision-makers, the model has allowed the introduction of demographic aspects of travellers into a network analysis process. Results showed a considerable difference between the flow patterns predicted by the proposed model and the traditional models.

Random Choice as Behavioral Optimization

We develop an extension of Luce's random choice model to study violations of the weak axiom of revealed preference. We introduce the notion of a stochastic preference and show that it implies the Luce model. Then, to address well-known difficulties of the Luce model, we define the attribute rule and establish that the existence of a well-defined stochastic preference over attributes characterizes it. We prove that the set of attribute rules and random utility maximizers are essentially the same. Finally, we show that both the Luce and attribute rules have a unique consistent extension to dynamic problems.

Optimization of Green Low Carbon Packaging Based on the Extreme Value and Color Fading Period

To construct low-carbon city, the development trend of green living is gradually clear, low-carbon has become the new trend of various fields development, and green planning, green building and low-carbon consumption also have become the mainstream of society. From the green low-carbon and minimalistic design of the packaging field starting, first expounded the importance of packaging materials green low-carbon and its superiority analysis, and then the use of random choice model makes science choice analysis of packaging materials, further using the deterministic model analyzes the layout and color of packaging design, to provide the theoretical basis and practical new model for green low-carbon packaging and minimalistic design.

Consumer preferences for alternative fuel vehicles: Comparing a utility maximization and a regret minimization model

This paper presents a utility-based and a regret-based model of consumer preferences for alternative fuel vehicles, based on a large-scale stated choice-experiment held among company car leasers in The Netherlands. Estimation and application of random utility maximization and random regret minimization discrete choice models shows that while the two models achieve almost identical fit with the data and differ only marginally in terms of predictive ability, they generate rather different choice probability-simulations and policy implications. The most eye-catching difference between the two models is that the random regret minimization model accommodates a compromise-effect, as it assigns relatively high choice probabilities to alternative fuel vehicles that perform reasonably well on each dimension instead of having a strong performance on some dimensions and a poor performance on others.

Do hens have friends?

Recent interest in positive welfare has encouraged consideration of the formation of socio-positive relationships in farmed species which may provide a means by which to manage positive states. We investigated in detail the existence of dyadic preferential associations in small groups of domestic laying hens. Spatial and temporal associations were examined in two contexts (day activity and evening roosting), within 8 identical pens of 15 laying hens over 8 weeks. Little aggression was observed. Social network analysis was performed to investigate correlations in who associated with whom using weighted degree (number) and binary (presence or absence) data for shared resource areas and proximity to other individuals. No consistent evidence was found for hens actively preferring others in their choice of resource area, or in companion proximity. Perch-roosting positions chosen by the hens were compared with data generated from a random-choice model. Hens showed no position preferences. Most dyads were never observed roosting together and, although some apparently perched together frequently, the low number of nights perching and proportion of nights spent together indicates these findings should be interpreted with caution. Overall, we found no convincing evidence of dyadic preferential relationships expressed by close active and resting proximities. Further work is required to confirm whether these findings hold true in other experimental contexts, are affected by social experience and if they hold in common with the progenitor sub-species.

Intraguild predation in winter wheat: prey choice by a common epigeal carabid consuming spiders

Summary Predators can provide a valuable ecosystem service by suppressing crop pests. However, intraguild predation, where predators compete for the same prey resource whilst consuming each other, may destabilize population dynamics and increase the risk of pest outbreaks. Very little is known about intraguild predation in open fields or the strengths of trophic links between predators which may negatively affect pest control. We tested the null hypothesis that predation by the epigeal predator Pterostichus melanarius (Coleoptera: Carabidae) on different spiders is species‐independent (proportional to density). A combination of population monitoring in winter wheat, molecular identification of juvenile spiders, molecular analysis of predator gut contents and a Monte Carlo simulation model were used to analyse prey choice. Pterostichus melanarius were pitfall‐trapped over three months, and 622 individuals were screened for the remains of four spider species. Predation rates on spiders were 43·6% in June and 33·3% in August and showed clear evidence of prey choice. Predation on the web‐dependent Tenuiphantes tenuis (Linyphiidae) was significantly greater than predicted from a random choice model, while predation on Bathyphantes gracilis (Linyphiidae) was significantly lower. The beetles may be selecting the most abundant species disproportionately (switching) or responding in some cases to spatial niche separation (T. tenuis locate their webs marginally lower than B. gracilis). However, two itinerant hunters, Erigone spp. (Linyphiidae) and Pachygnatha degeeri (Tetragnathidae), were consumed in proportion to their density. Synthesis and applications. High levels of intraguild predation were revealed using molecular diagnostics. The gut analysis approach provided invaluable data that will inform the future design of appropriate pest management and integrated farming strategies that encourage these predators. The data showed strong evidence of prey choice. Managers can, however, probably encourage high densities of all these known aphid predators (spiders and carabids) because disproportionately high rates of predation on the most common spiders at our field sites (T. tenuis) were not sufficient to prevent strong growth in the density of this species between June and August (adults increased × 1·6 and juveniles × 8·6). Such work is essential if we are to reveal the processes behind functional biodiversity in crops.

Modeling Automobile Drivers’ Toll-Lane Choice Behavior at a Toll Plaza

AbstractA toll-lane selection process at a toll plaza can be looked at as an outcome of the choice processes of individual drivers. In this paper, an attempt is made to develop a random utility-based discrete multinomial choice model for the behavior of automobile drivers while selecting toll lanes at a toll plaza. Specifically, a multinomial logit model is developed and calibrated using disaggregate level choice data from three toll plazas with different geometry and rates of arrival of vehicles. The calibrated logit models from the different sites when statistically compared show that a generic model, applicable to all the sites, is possible. Such a generic model is also developed. The use of the proposed model can improve the analysis of traffic flow at toll plazas and ultimately lead to more effective designs of these facilities.

How we Became the Schmams: Conceptualizations of Fairness in the Decision‐Making Process for Latina/O Children

The current study examines 16 Latina/o fifth grade children's desires for a decision-making structure within a youth participatory action research (yPAR) program. When given the choices of consensus, majority rule, authoritarian rule, delegation, and random choice models, children chose random choice. Procedural, distributive and emotional justice were heavily weighted in their reasoning around fairness and decision making. Many thought random choice offered the best alternative because it flattened power hierarchies so that each child would, at some point, have the power to make a decision. Additionally, children argued that the neutrality of random choice allowed them to sidestep interpersonal tensions. Implications include how social identities inform definitions of fairness and how yPAR programs should work with youth around how they will make decisions.

VOLATILITY IN THE CONSUMER PACKAGED GOODS INDUSTRY — A SIMULATION BASED STUDY

The volatility in a CPG market is modeled using a bottom-up simulation approach and validated against disaggregated supermarket transactions data. The simulation uses independent agents, each agent representing unique households in the data. A simple behavioral model incorporates household preferences for product attributes and prices. Our validation strategy tests the model predictions at both macro and micro levels and benchmarks the performance in each against a random choice model. The model significantly outperforms the benchmark at both levels. At the macro level, choices made by heterogenous agents accurately captures the volatility in market shares over time. This accuracy at the macro level is driven by the accuracy of predictions at the micro household level SKU and attribute choice.

Volatility in the Consumer Packaged Goods Industry: A Simulation Based Study

The volatility in a CPG market is modeled using a bottom up simulation approach and validated against disaggregated supermarket transactions data. The simulation uses independent agents, each agent representing unique households in the data. A simple behavioral model incorporates household preferences for product attributes and prices. Our validation strategy tests the model predictions at both macro and micro levels and benchmarks the performance in each against a random choice model. The model significantly outperforms the benchmark at both levels. At the macro level, choices made by heterogenous agents accurately captures the volatility in market shares - with the direction of change being more accurately predicted than the magnitude. At the micro level, it achieves a reasonable degree of prediction accuracy of household level SKU choice and a substantially higher accuracy for attribute choice.

On the Recoverability of Choice Behaviors with Random Coefficients Choice Models in the Context of Limited Data and Unobserved Effects

Random coefficients choice models are seeing widespread adoption in marketing research, partly because of their ability to generate household-level parameter estimates with limited data. However, the power of such models may tempt researchers to trust that they continue to produce reasonable estimates, when in fact either model misspecification or insufficient data limits the models' ability to recover household-level parameters successfully. If household-level choice behaviors are not recovered successfully, managerial decisions such as marketing-mix planning and targeting, direct marketing, segmentation, and forecasting may not produce the desired results. This study addresses the following questions. First, can random coefficients choice models correctly identify markets characterized by preference and response heterogeneity, state dependence, the use of alternative decision heuristics that result in reduced choice sets, and combinations of these effects? If so, how much data is required, and is this realistic given the size of data sets typically used in marketing analyses? Which model selection criteria should be used to identify these markets? When there is spurious market identification, which parameters contribute to the spurious result? An extensive simulation experiment is conducted wherein random coefficients logit models with varying specifications of parameter heterogeneity, state dependence effects, and choice set heterogeneity are applied to 128 experimental conditions. The results show which types of markets can be identified reliably and which cannot. Based on the results of the simulation, the authors develop a model selection heuristic that identifies the correct market in 81% of the experimental conditions. In contrast, strict application of the best model selection criterion alone results in correct market identification in at most 34% of experimental conditions. Interestingly, we find that the amount of data (number of households or number of purchases per household) does not affect our ability to identify the correct market type with this heuristic, so there is a good chance of identifying the correct market type even when little data is available.

Random effects choice models: seeking latent predisposition segments in the context of retail store format selection

The aim of this paper is to gain insight into the potential use of random effects choice models as analytical instruments. These instruments then may be used to identify the latent predisposition and benefit segments of a market in the context of store selection, specifically, in the context of the selection of store format. First, the theoretical bases, formulations, and practical aspects that make up the modelling technique in question are reviewed and revised. Second, an adaptation to detect latent segments with predispositions to shop at a store format, isolated from spatial coverage factors, is proposed. Third, the proposal is empirically tested in the case of the hypermarket. The results confirm the explanatory possibilities of this modelling approach and show that various segments with different levels of response to the hypermarket exist. They can be characterised ad hoc by means of demographic and sociological variables to facilitate their identification, access, and evaluation, with a view to possible marketing strategies.

Commuters? non-linear response to time distances

Established analyses of labour market commuting are based on random choice models and gravity type models. In these models generalised transport costs are formulated as exponential or log-linear distance dependent functions. This paper presents empirical observations, which imply that time distances influence the commuting behaviour in a non-linear way, such that the time sensitivity is much lower for very short and long distances, whereas intermediate distances display a high time sensitivity. This is explained in a model that is parameterised and estimated. The results are important for understanding and predicting commuter behaviour. It also helps to delineate space, as in classical traditions, into local, intra-regional and extra-regional space.

Using stated preference discrete choice modelling to inform health care decision-making: A pilot study of breast screening participation

This study was an important start to explore the feasibility of applying stated preference discrete choice modelling (SPDCM) for use in developing breast screening participation enhancement strategies. It needs to be followed by further research to establish model validity and authoritative results. In the meantime a random effects binary probit choice model was estimated using a main effects with selected 2-way interaction design and a convenience sample of Australian breast cancer screening participants. A response rate of 48% was obtained. Clear preferences for different service configurations were revealed and used to demonstrate how potential strategies to enhance future participation rates of women placed on routine recall could be identified. As anticipated accuracy of screening was the most important attribute of the service to influence the probability of uptake but others were screening time, travel time, information about screening benefits and the desire for privacy lending support to the view that benefit assessment goes beyond health factors. In summary, the SPDCM approach can be regarded as a judicious approach for helping decision-makers improve screening participation.

The Value of Purchase History Data in Target Marketing

An important aspect of marketing practice is the targeting of consumer segments for differential promotional activity. The premise of this activity is that there exist distinct segments of homogeneous consumers who can be identified by readily available demographic information. The increased availability of individual consumer panel data open the possibility of direct targeting of individual households. The goal of this paper is to assess the information content of various information sets available for direct marketing purposes. Information on the consumer is obtained from the current and past purchase history as well as demographic characteristics. We consider the situation in which the marketer may have access to a reasonably long purchase history which includes both the products purchased and information on the causal environment. Short of this complete purchase history, we also consider more limited information sets which consist of only the current purchase occasion or only information on past product choice without causal variables. Proper evaluation of this information requires a flexible model of heterogeneity which can accommodate observable and unobservable heterogeneity as well as produce household level inferences for targeting purposes. We develop new econometric methods to implement a random coefficient choice model in which the heterogeneity distribution is related to observable demographics. We couple this approach to modeling heterogeneity with a target couponing problem in which coupons are customized to specific households on the basis of various information sets. The couponing problem allows us to place a monetary value on the information sets. Our results indicate there exists a tremendous potential for improving the profitability of direct marketing efforts by more fully utilizing household purchase histories. Even rather short purchase histories can produce a net gain in revenue from target couponing which is 2.5 times the gain from blanket couponing. The most popular current electronic couponing trigger strategy uses only one observation to customize the delivery of coupons. Surprisingly, even the information contained in observing one purchase occasion boasts net couponing revenue by 50% more than that which would be gained by the blanket strategy. This result, coupled with increased competitive pressures, will force targeted marketing strategies to become much more prevalent in the future than they are today.

Affinities and frictions of trade networks

This paper formulates a theoretical framework in which the analysis starts with supplier-customer links of micro type. From this we derive aggregate trade links between nodes in a spatial network. It is shown how affinities and barriers shape and influence the establishment of customer links, and how this forms aggregate patterns of flows. The framework is applied to three types of models for determining trade patterns — random discrete choice, information theory and gravity models. Essentially equivalent solutions are derived.

Propofol at a subanesthetic dose may have abuse potential in healthy volunteers.

We conducted a study in which the rewarding effects of propofol were assessed in humans, using a discrete-trials choice procedure. Normal healthy volunteers (n = 12) were exposed in a blind fashion to acute bolus injections of 0.6 mg/kg of propofol twice and to a similar volume of Intralipid twice. Then, for the next three sessions, subjects chose which drug (identified by a color code) they wished to receive. We defined propofol choosers if individuals chose propofol two or three times, and nonchoosers if they chose propofol once or not at all. By using a chi 2 goodness-of-fit test of a random choice model, the choice distribution differed significantly from a random choice distribution (P < 0.005). Six subjects were choosers: four subjects chose propofol on all three choice occasions and two subjects chose the drug on 2/3 occasions (referred to hereafter as propofol choosers). Six subjects were nonchoosers: five subjects chose Intralipid on all three choice occasions, and one subject chose Intralipid twice (referred to hereafter as propofol nonchoosers). During sampling sessions, propofol choosers reported pleasant acute effects and no unpleasant residual effects, whereas propofol nonchoosers reported either unpleasant acute subjective effects and/or residual effects from propofol. We conclude that 1) propofol may be rewarding (reinforcing) in some individuals without a history of drug abuse, and 2) further abuse liability testing is needed with this drug.