nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2023‒09‒25
nine papers chosen by
Edoardo Marcucci, Università degli studi Roma Tre

  1. The Role of Respondent Certainty and Attribute Non-Attendance on the Willingness to Pay for the Attributes of Recyclable Aluminum Bottled Water By Paul Hindley; O. Ashton Morgan
  2. Farmer-friendly delivery of veterinary services: Experimental insights from the Kenyan dairy sector By Maina, Kevin W.; Parlasca, Martin C.; Rao, Elizaphan J.O.; Qaim, Matin
  3. Bootstrap inference for fixed-effect models By Koen Jochmans
  4. New approaches to measuring welfare By Cooper, Kristen; Fabian, Mark; Krekel, Christian
  5. On the ratios of urban mobility, Part 1: the HoTer model of travel demand and network flows By Fabien Leurent
  6. Choice Architecture, Privacy Valuations, and Selection Bias in Consumer Data By Tesary Lin; Avner Strulov-Shlain
  7. Optimal Shrinkage Estimation of Fixed Effects in Linear Panel Data Models By Soonwoo Kwon
  8. Scalable Estimation of Multinomial Response Models with Uncertain Consideration Sets By Siddhartha Chib; Kenichi Shimizu
  9. External Validity of Inferred Attribute NonAttendance: Evidence from a Laboratory Experiment with Real and Hypothetical Payoffs By Tanga Mohr; John C. Whitehead

  1. By: Paul Hindley; O. Ashton Morgan
    Abstract: With the recycling constraints on traditional plastic bottles and environmental concerns regarding the volume of non-recycled plastic packaging, aluminum bottles and cans offer an environmentally-friendly alternative to packaging drinking water. This research utilizes a stated preference discrete choice experiment to measure consumers’ willingness to pay for recyclable aluminum water bottles and their attributes. We find that the type of bottle top is crucial, with consumers willing to pay a premium for resealable aluminum water bottles compared to a plastic bottles but more for plastic bottles over aluminum cans with a non-resealable pop top. This provides insight into the potential for using recycled aluminum packaging in bottled water production to mitigate the volume of plastics in the environment. The application also examines model calibration to address choice certainty and inferred attribute non-attendance. Our findings also indicate that accounting for choice certainty and inferred attribute non-attendance can influence attribute coefficient estimates and marginal willingness to pay. Key Words: willingness to pay, certainty, attribute non-attendance, discrete choice experiment
    Date: 2023
  2. By: Maina, Kevin W.; Parlasca, Martin C.; Rao, Elizaphan J.O.; Qaim, Matin
    Abstract: Poor health conditions of livestock cause sizeable losses for many farmers in the Global South. Veterinary services, including vaccinations, could help but often fail to reach farmers under typical smallholder conditions. Here, we examine how the provision of a vaccine against East Cost Fever (ECF) – a tick-borne disease affecting cattle in Africa – can be designed to reduce typical adoption barriers. Using data from a choice experiment with dairy farmers in Kenya, we evaluate farmers’ preferences and willingness to pay for various institutional innovations in vaccine delivery, such as a stronger role of dairy cooperatives, new payment modalities with a check-off system, vaccination at farmers’ homestead, and bundling vaccinations with discounts for livestock insurance. Our data reveal that farmers’ awareness of the ECF vaccine is limited and adoption rates are low, largely due to institutional constraints. Results from mixed logit and latent class models suggest that suitable institutional innovations – tailored to farmers’ heterogeneous conditions – could significantly increase adoption. This general finding likely also holds for other veterinary technologies and services in the Global South.
    Keywords: Farm Management, Livestock Production/Industries
    Date: 2023–09–07
  3. By: Koen Jochmans (École d'économie de Toulouse)
    Abstract: The maximum likelihood estimator of nonlinear panel-data models with fixed effects is asymptotically biased under rectangular-array asymptotics. The literature has devoted substantial effort to devising methods that correct for this bias as a means to salvage standard inferential procedures. The chief purpose of this presentation is to show that the (recursive, parametric) bootstrap replicates the asymptotic distribution of the (uncorrected) maximum-likelihood estimator and of the likelihood-ratio statistic. This justifies the use of confidence sets and decision rules for hypothesis testing constructed via conventional bootstrap methods. No modification for the presence of bias needs to be made.
    Date: 2023–08–11
  4. By: Cooper, Kristen; Fabian, Mark; Krekel, Christian
    Abstract: Economics has traditionally understood ‘welfare’ (what makes a life go well) as the satisfaction of preference. This conceptualisation of welfare is typically measured using revealed preferences, proxied through income and prices or stated in willingness-to-pay surveys. Recent decades have seen growing challenges to this paradigm. The climate crisis, among other phenomena, has called into question whether income and price data effectively proxy preferences, and willingness-to-pay surveys continue to struggle with accurately pricing important items such as biodiversity, digital goods, privacy and social connections. Preference satisfaction as a welfare criterion has also been challenged conceptually by psychologists and scholars working in the development space, among others. In this article, we review recent innovations in alternate ways of conceptualising and measuring welfare for the purposes of economic welfare analysis. We focus on using stated preferences over aspects of well-being, life-satisfaction scales and the WELLBY approach, and well-being frameworks such as Bhutan's Gross National Happiness Index. While not without weaknesses, these approaches also have marked strengths relative to the traditional approach.
    JEL: J1
    Date: 2023–08–24
  5. By: Fabien Leurent (CIRED - Centre International de Recherche sur l'Environnement et le Développement - Cirad - Centre de Coopération Internationale en Recherche Agronomique pour le Développement - EHESS - École des hautes études en sciences sociales - AgroParisTech - ENPC - École des Ponts ParisTech - Université Paris-Saclay - CNRS - Centre National de la Recherche Scientifique)
    Abstract: Mobility systems in urbanized territories have been featured out in Travel Demand Models by state variables of land-use occupation, trip generation, trip distribution, modal split and network assignment, with emphasis on causal relationships between the variables and on spatial detail for each kind of variables. The article is aimed to provide notional averages, say ratios, for each kind of variables, and to state the causal relationships between the variables as simple analytical formulas between the ratios. This is achieved by going along the classical four steps of travel demand modeling, in a theoretical way for an idealized territory satisfying three postulates of homogeneity: namely, at block level, at link level and of indefinite spatial extension. The said formulas constitute rules of thumb linking the mobility ratios of spatial density of human occupation, trip emission rates, average trip lengths, modal shares, generalized trip cost per length unit, together with traffic variables of speed, flow rate and vehicular density at the link level. The model is stated in eight steps, namely (i) territorial composition, (ii) trip generation, (iii) trip lengths and traffic formation, (iv) quality of service, (v) trip distribution using a gravity model, (vi) modal split by multinomial logit, (vii) traffic laws, (viii) traffic equilibrium. It is followed by a Discussion of the model outreach and limitations. Areas of further research include traffic laws, impact assessment and economic analysis.
    Keywords: Spatial homogeneity, State laws, Four-step travel demand model, Traffic equilibrium Highlights
    Date: 2022–10–12
  6. By: Tesary Lin; Avner Strulov-Shlain
    Abstract: We study how choice architecture that companies deploy during data collection influences consumers' privacy valuations. Further, we explore how this influence affects the quality of data collected, including both volume and representativeness. To this end, we run a large-scale choice experiment to elicit consumers' valuation for their Facebook data while randomizing two common choice frames: default and price anchor. An opt-out default decreases valuations by 14-22% compared to opt-in, while a \$0-50 price anchor decreases valuations by 37-53% compared to a \$50-100 anchor. Moreover, in some consumer segments, the susceptibility to frame influence negatively correlates with consumers' average valuation. We find that conventional frame optimization practices that maximize the volume of data collected can have opposite effects on its representativeness. A bias-exacerbating effect emerges when consumers' privacy valuations and frame effects are negatively correlated. On the other hand, a volume-maximizing frame may also mitigate the bias by getting a high percentage of consumers into the sample data, thereby improving its coverage. We demonstrate the magnitude of the volume-bias trade-off in our data and argue that it should be a decision-making factor in choice architecture design.
    Date: 2023–08
  7. By: Soonwoo Kwon
    Abstract: Shrinkage methods are frequently used to estimate fixed effects to reduce the noisiness of the least square estimators. However, widely used shrinkage estimators guarantee such noise reduction only under strong distributional assumptions. I develop an estimator for the fixed effects that obtains the best possible mean squared error within a class of shrinkage estimators. This class includes conventional shrinkage estimators and the optimality does not require distributional assumptions. The estimator has an intuitive form and is easy to implement. Moreover, the fixed effects are allowed to vary with time and to be serially correlated, and the shrinkage optimally incorporates the underlying correlation structure in this case. In such a context, I also provide a method to forecast fixed effects one period ahead.
    Date: 2023–08
  8. By: Siddhartha Chib; Kenichi Shimizu
    Abstract: A standard assumption in the fitting of unordered multinomial response models for J mutually exclusive nominal categories, on cross-sectional or longitudinal data, is that the responses arise from the same set of J categories between subjects. However, when responses measure a choice made by the subject, it is more appropriate to assume that the distribution of multinomial responses is conditioned on a subject-specific consideration set, where this consideration set is drawn from the power set of {1, 2, ..., J}. Because the cardinality of this power set is exponential in J, estimation is infeasible in general. In this paper, we provide an approach to overcoming this problem. A key step in the approach is a probability model over consideration sets, based on a general representation of probability distributions on contingency tables. Although the support of this distribution is exponentially large, the posterior distribution over consideration sets given parameters is typically sparse, and is easily sampled as part of an MCMC scheme that iterates sampling of subject-specific consideration sets given parameters, followed by parameters given consideration sets. The effectiveness of the procedure is documented in simulated longitudinal data sets with J=100 categories and real data from the cereal market with J=73 brands.
    Date: 2023–08
  9. By: Tanga Mohr; John C. Whitehead
    Abstract: We consider differences in hypothetical and real payoff laboratory experiments using attribute non-attendance methods. Attribute non-attendance is an empirical approach that measures and accounts for when survey respondents ignore attributes in stated preference surveys. We use attribute non-attendance methods with data from an emissions permit experiment with real and hypothetical payments. Our conjecture is that attribute non-attendance may be more pronounced in hypothetical sessions and, once accounted for, hypothetical decisions and real decisions influenced by monetary payoffs will be more similar. In both treatments we find that the effect of the cost of an emissions permit on behavior differs if the cost is implicit or explicit. In inferred attribute non-attendance models with the real treatment data we find two classes of respondents with different behavior but no evidence of attribute non-attendance. With the hypothetical treatment data we find two classes of respondents with different behavior and evidence of attribute non-attendance on two of the four choice attributes. Key Words: attribute non-attendance, emissions permits, laboratory experiment, stated preferences
    Date: 2023

This nep-dcm issue is ©2023 by Edoardo Marcucci. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.