RUM Model Research Articles

A Zonal RUM Model to Value Recreation Sites with Aggregate Visitation Data

A Zonal RUM Model to Value Recreation Sites with Aggregate Visitation Data

Stimulus Perception in Long-Distance Railway Mode Choice

In the context of countries in the so-called Global South, where passenger railway services are either nonexistent or poorly performed, discrete choice models are useful to identify the attributes that affect users’ choices and provide insights on their behaviour in regional long-distance trips. Several theories and models have been proposed to understand travel behaviour for effective strategical decision in the transport field. The well-knownRandom Utility Maximization (RUM) approach has been widely used for such purposes, while the Random Regret Minimization (RRM) approach has been recently explored in the literature. However, the magnitude in the difference of levels of the attributes, or the stimulus perception, may affect the results of such models and biases the estimations. Therefore, this paper aims to assess the stimulus perception in mode choice to compare conventional rail (CR) and high-speed rail (HSR) services for passenger transport in intercity trips in Brazil. Estimations of RUM and RRM models were performed with a dataset from a stated preference survey comparing two railway technologies (CR and HSR) with other modes of transport (car, bus, and airplanes) for long-distance trips in the Southeast region of Brazil. Findings provide useful insights about the impacts of travel costs, travel times, and frequency of services, as well as sociodemographic characteristics of users. From the modelling outputs, it was found that users are affected by the magnitude of travel costs, time, and frequency only in business trips by HSR in the Brazilian context.

RED: RFID-Based Eccentricity Detection for High-Speed Rotating Machinery

Eccentricity detection is a crucial issue for high-speed rotating machinery, which concerns the stability and safety of the machinery. Conventional techniques in industry for eccentricity detection are mainly based on measuring certain physical indicators, which are costly and hard to deploy. In this paper, we propose RED, a non-intrusive, low-cost, and real-time RFID-based eccentricity detection approach. Differing from the existing RFID-based sensing approaches, RED utilizes the temporal and phase distributions of tag readings as effective features for eccentricity detection. RED includes a Markov chain based model called RUM, which only needs a few sample readings from the tag to make a highly accurate and precise judgement. The design of RED further addresses practical issues, such as parameterizing the RUM model, making it robust to dynamic and noisy environments, and considering how the doppler shift may affect our system. We implement RED with COTS RFID reader and tags, and evaluate its performance across various scenarios. The overall accuracy is 93.6 percent and the detection latency is 0.68 seconds in average.

Choice probabilities of random utility maximization models when the errors distribution is a polynomial copula with Gumbel marginals

ABSTRACT The expression of the choice probabilities of RUM models in which the utility errors distribution is a polynomial copula with Gumbel marginals is derived. In particular, it is shown that the expression of the Choice Probability Generating Function can be easily obtained from the polynomial copula of the distribution. This result is particularized for two types of polynomial copulas: the multivariate Farlie–Gumbel–Morgenstern copulas and the Baker's copulas. The correlation coefficient of the errors with these copulas is also obtained. The corresponding RUM models may account for relatively large negative and positive correlation. The RUM models introduced in this work are applied to two samples of interurban trips with three alternatives: airplane, rail and car. The results of the fit show that the models presented in this work may account for the correct correlation signbetween the alternative errors and yield a better fit than a nested logit model.

Preferences of selecting public transit for inter-city passengers using the stated preference method with RUM and RRM Model

At four step model analysis there are two model choice must be known, a vehicle choice mode (mode choice), and the selection of the route (Route Choice). So far, the model developed in the world of transportation (RUM) has been based on economic theory, which emphasizes rationality in an election. Regret theory (RRM) developed a theory of voter behavior (choice behavior), where there is a state of uncertainty in it. Regret theory is used in various disciplines such as marketing, micro economy, psychology, management, and transportation. This study aims to demonstrate differences in the results of calculation of the analysis (RUM and RRM), in the case of choosing the type of vehicle, with Medan to Balige route, North Sumatera, where there are two alternative modes of vehicles, with three utilities in each of these alternatives. The results showed a significant difference in the results of the developed model.

Is your dataset big enough? Sample size requirements when using artificial neural networks for discrete choice analysis

Artificial Neural Networks (ANNs) are increasingly used for discrete choice analysis. But, at present, it is unknown what sample size requirements are appropriate when using ANNs in this particular context. This paper fills this knowledge gap: we empirically establish a rule-of-thumb for ANN-based discrete choice analysis based on analyses of synthetic and real data. To investigate the effect of complexity of the data generating process on the minimum required sample size, we conduct extensive Monte Carlo analyses using a series of different model specifications with different levels of model complexity, including RUM and RRM models, with and without random taste parameters. Based on our analyses we advise to use a minimum sample size of fifty times the number of weights in the ANN; it should be noted, that the number of weights is generally much larger than the number of parameters in a discrete choice model. This rule-of-thumb is considerably more conservative than the rule-of-thumb that is most often used in the ANN community, which advises to use at least ten times the number of weights.

On the robustness of random regret minimization modelling outcomes towards omitted attributes

As discrete choice models may be misspecified, it is crucial for choice modellers to have knowledge on the robustness of their modelling outcomes towards misspecification. This study investigates the robustness of Random Regret Minimization (RRM) modelling outcomes towards one sort of model misspecification: the omission of relevant attributes. We explore the effect of omitted attributes (orthogonal and correlated) in the context of labelled and unlabeled data. In the context of labelled data, we show that – just as in RUM models – in RRM models Alternative Specific Constants (ASCs) can be used to capture the average effect of omitted attributes. However, in contrast to RUM models, ASCs in RRM models are choice set composition specific. This implies that in order to achieve consistent parameter estimates when the choice set composition varies across choice observations, different sets of ASCs need to be estimated for each unique choice set composition. In the context of unlabeled data, we show – using Monte Carlo simulations – that RRM models are fairly robust towards the presence of an orthogonal omitted attribute, though not as robust as the linear-additive RUM model. Specifically, we find that: (1) Aggregate Demand Elasticities (ADEs) implied by RRM models are less robust towards the presence of an orthogonal omitted attributes than those implied by linear-additive RUM models, and (2) Average Sample Effects (ASEs) implied by RRM models are – in the presence of an omitted orthogonal attribute – more sensitive towards misspecification in terms of the underlying decision rule than those implied by its linear-additive RUM counterpart.

The effect of choice set misspecification on welfare measures in random utility models

Random utility models have been widely employed in environmental valuation. But stochastic choice set formation models in the random utility framework are rarely applied in this literature, although previous research has shown that ignoring choice set formation (when it exists) leads to biased parameter estimates and welfare measures. This paper conducts Monte Carlo (MC) experiments to investigate the directionality and magnitude of welfare measure bias arising from ignoring or misspecifying choice set formation. We find that when attribute cutoff-based choice set formation exists in the data generation process, typical RUM models ignoring or misspecifying choice set formation underestimate welfare measures by 30–50%. Models that approximate choice set formation may produce unbiased welfare measures, but constitute a promising area for future research. This paper illustrates the importance of applying choice set formation in environmental valuation and provides practical guidance about the usage of stochastic choice set formation models.

New insights on random regret minimization models

This paper develops new methodological insights on Random Regret Minimization (RRM) models. It starts by showing that the classical RRM model is not scale-invariant, and that – as a result – the degree of regret minimization behavior imposed by the classical RRM model depends crucially on the sizes of the estimated taste parameters in combination with the distribution of attribute-values in the data. Motivated by this insight, this paper makes three methodological contributions: (1) it clarifies how the estimated taste parameters and the decision rule are related to one another; (2) it introduces the notion of “profundity of regret”, and presents a formal measure of this concept; and (3) it proposes two new family members of random regret minimization models: the μRRM model, and the Pure-RRM model. These new methodological insights are illustrated by re-analyzing 10 datasets which have been used to compare linear-additive RUM and classical RRM models in recently published papers. Our re-analyses reveal that the degree of regret minimizing behavior imposed by the classical RRM model is generally very limited. This insight explains the small differences in model fit that have previously been reported in the literature between the classical RRM model and the linear-additive RUM model. Furthermore, we find that on 4 out of 10 datasets the μRRM model improves model fit very substantially as compared to the RUM and the classical RRM model.

The size of the cliff at the border

The scale of international migration flows depends on moving costs that are, in turn, influenced by host-country policies. This paper shows how to estimate the influence of policies upon bilateral migration flows to multiple destinations. We rely on a Poisson pseudo-maximum likelihood estimator to derive estimates that are consistent under more general distributional assumptions on the underlying RUM model than the ones commonly adopted in the literature. We derive bounds for the estimated direct and indirect effects of migration policies that reflect the uncertainty connected to the use of aggregate data, and we show that bilateral migration flows can be highly sensitive to the immigration policies set by other destination countries, an externality that we are able to quantify.

Toward an Understanding of the Local Origin of Negative Thermal Expansion in ZrW2O8: Limits and Inconsistencies of the Tent and Rigid Unit Mode Models

Although zirconium tungstate (ZrW2O8) is the most popular negative thermal expansion (NTE) material, the exact mechanism responsible for its NTE still remains controversial. Specifically, the “Tent” model [Cao et al. Phys. Rev. Lett. 2002, 89, 215902; Bridges et al. Phys. Rev. Lett. 2014, 112, 045505] and the “rigid unit mode” (RUM) model [Tucker et al. Phys. Rev. Lett. 2005, 95, 255501] were subjects of debate during recent years. This work aims to shed light on this issue by means of molecular dynamics simulations which allow us to separate, for each bond distance, the “true” thermal expansion from the “apparent” thermal expansion, as well as to study the effective bond strength and the anisotropy of relative thermal motion. In spite of the good agreement with the experimental data of Cao, Bridges, and co-workers, a decrease of the “true” W–Zr distances has been observed accompanied by large transverse vibrations of the O atoms in the middle of the W–O–Zr linkage, in sharp contrast to the “tent” model. Moreover, in contrast to the RUM model, it has been found that the WO4 and ZrO6 polyhedra are strongly distorted by thermal motion, and, more importantly, that intra-polyhedra contributions to NTE are present. Accordingly, we can conclude that both the tent and RUM models are inadequate to explain NTE in ZrW2O8, and a more flexible model, simply based on rigid nearest W–O and Zr–O bonds and tension effect, should be adopted.

Relative advantage maximisation as a model of context dependence for binary choice data

In recent years we have seen a burgeoning literature, within a discrete choice modelling setting, focussing on capturing increased behavioural realism in decision making through the use of alternative specifications of the utility function. One notable example is the increasing popularity of the random regret minimisation (RRM) model, in which respondents are assumed to choose the alternative that minimises the negative emotions associated with decision making.In this paper, we discuss a ‘relative advantage maximisation’ (RAM) model, which shares some similar characteristics to the RRM model in terms of how contextual effects in the choice set are accounted for. The key feature of the RAM model is the relative advantage component which frames the comparison of an attribute against its counterparts in all other alternatives in the choice set as either an advantage or a disadvantage. The relative advantage is so named because of the assumption that it is the advantage of an alternative relative to the sum of its advantage and disadvantage that matters in the utility function.Although the RRM model has been successfully applied to choice sets with three or more alternatives, with binary choice data, one significant limitation to its scope is that the RRM model collapses to the standard RUM model. This is not the case with the RAM model, and so the notion of context dependency is still meaningfully preserved under RAM. Using the MNL model, we find that the RAM model performs no worse than the RUM model in all six binary choice datasets analysed. In some cases, a small but significant improvement in model fit is observed. However, more advanced models such as the error components and random parameters logit present a more empirically mixed picture for the RAM model. Marginal willingness to pay (WTP) measures derived from the RAM model are also discussed. Differences may be negligible in the MNL case, but not necessarily so for the random parameter logit.

A Random Parameter Model with Onsite Sampling for Recreation Site Choice: An Application to Southern California Shoreline Sportfishing

Estimation of consistent parameter estimates for recreational demand models faces challenges arising from the choice-based nature of the data collected primarily for resource management purposes. As an alternative to randomized respondent-based sampling, choice-based onsite sampling can provide information on actual choices made by a subset of the population where participation has a low incidence. While the literature has shown that under specific restrictions the estimation of choice models from onsite sampling data yields unbiased fixed parameter estimates for the conditional logit model, this result does not carry over to estimation of the random parameter logit model. We propose an estimator for the unbiased estimation of the random parameter model using choice-based data; our estimator uses weights based on information about the level of sampling effort. An empirical application of the standard and weighted discrete choice RUM models to onsite sample data on recreational fishing illustrates the advantages of the proposed estimator. The estimation results indicate the compensating variation associated with an decrease, or increase, of 50 % in expected catch rates for a recreational shoreline sportfishing trip to a man-made structure in southern California is $$-{\$}2.80$$ or $${\$}3.54$$ per trip, respectively.

Controlling for Observed and Unobserved Site Characteristics in RUM Models of Recreation Demand

Recreation demand models are typically plagued by limited information on site attributes. If these unobserved site attributes are correlated with the observed characteristics and/or the travel cost variable, the resulting parameter estimates are likely to be biased. We develop a Bayesian approach to estimating a model that incorporates a full set of alternative‐specific constants, insulating the key travel cost parameter from the influence of unobservables. The proposed posterior simulator can be used in the mixed logit framework in which some parameters of the conditional utility function are random. We apply the estimation procedures to data from the Iowa Lakes Project.

High-pressure thermo-elastic properties of beryl (Al4Be6Si12O36) from ab initio calculations, and observations about the source of thermal expansion

Ab initio calculations of thermo-elastic properties of beryl (Al4Be6Si12O36) have been carried out at the hybrid HF/DFT level by using the B3LYP and WC1LYP Hamiltonians. Static geometries and vibrational frequencies were calculated at different values of the unit cell volume to get static pressure and mode-γ Gruneisen’s parameters. Zero point and thermal pressures were calculated by following a standard statistical-thermodynamics approach, within the limit of the quasi-harmonic approximation, and added to the static pressure at each volume, to get the total pressure (P) as a function of both temperature (T) and cell volume (V). The resulting P(V, T) curves were fitted by appropriate EoS’, to get bulk modulus (K 0) and its derivative (K′), at different temperatures. The calculation successfully reproduced the available experimental data concerning compressibility at room temperature (the WC1LYP Hamiltonian provided K 0 and K′ values of 180.2 Gpa and 4.0, respectively) and the low values observed for the thermal expansion coefficient. A zone-centre soft mode $$ P6/mcc \to P\bar{1} $$ phase transition was predicted to occur at a pressure of about 14 GPa; the reduction of the frequency of the soft vibrational mode, as the pressure is increased, and the similar behaviour of the majority of the low-frequency modes, provided an explanation of the thermal behaviour of the crystal, which is consistent with the RUM model (Rigid Unit Model; Dove et al. in Miner Mag 59:629–639, 1995), where the negative contribution to thermal expansion is ascribed to a geometric effect connected to the tilting of rigid polyhedra in framework silicates.

Dynamic structural disorder incristobalite: neutron total scattering measurement and reverseMonte Carlo modelling

The atomic structure of cristobalite in both its high-temperatureβ-phase and low-temperature α-phase has been studied usingconstrained reverse Monte Carlo modelling of neutron totalscattering data. The modelling has shown that the disorder of theβ-phase involves rotations and displacements of rigid SiO4tetrahedra, consistent with the predictions of the `rigid-unit mode'(RUM) model. The structure of the α-phase is significantlydifferent from that of the β-phase, but still has a degree ofdisorder. The calculated distribution of oxygen atoms inβ-cristobalite is continuous rather than based on specificcrystallographic sites. There is no evidence for correlationsbetween the Si-O distance and the corresponding Si-O-Si bondangles, contrary to quantum mechanical calculation findings. Thethree-dimensional diffuse scattering has been calculated from themodel configurations, and is found to be in agreement withexperimental electron diffraction measurements and the predictionsof the RUM model. There is little significant temperaturedependence of the structure or diffuse scattering in the β-phase.