nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2013‒06‒09
ten papers chosen by
Edoardo Marcucci
Universita' di Roma Tre

  1. Best and worst choices By André De Palma; Karim Kilani; Gilbert Laffond
  2. Testing for Time-Invariant Unobserved Heterogeneity in Generalized Linear Models for Panel Data By Francesco Bartolucci; Federico Belotti; Franco Peracchi
  3. You can't always get what you want - East and West Germans' attitudes and preferences regarding the welfare state By Pfarr, Christian; Schmid, Andreas; Ulrich, Volker
  4. Generalised empirical likelihood-based kernel density estimation By Vitaliy Oryshchenko; Richard J. Smith
  5. How consumers’ socio-economic background influences satisfaction: Insights for better utility regulation By Clifton, Judith; Díaz-Fuentes, Daniel; Fernández-Gutiérrez, Marcos
  6. Ranking Scientific Journals via Latent Class Models for Polytomous Item Response By Francesco Bartolucci; Valentino Dardanoni; Franco Peracchi
  7. Risk and time preferences under the threat of background risk: a case-study of lahars risk in central Java By Marc Willinger; Mohamed Ali Bchir; Carine Heitz
  8. The Taylor Decomposition: A Unified Generalization of the Oaxaca Method to Nonlinear Models By Stephen Bazen; Xavier Joutard
  9. Do Preferences for Job Attributes Provide Evidence of 'Hierarchy of Needs' By Cem BaÅŸlevent; Hasan KirmanoÄŸlu
  10. Likelihood inference in some finite mixture models By Xiaohong Chen; Maria Ponomareva; Elie Tamer

  1. By: André De Palma (ENS Cachan - Ecole Normale Supérieure de Cachan - École normale supérieure de Cachan - ENS Cachan); Karim Kilani (LIRSA, Conservatoire National des Arts et Métiers - Conservatoire National des Arts et Métiers (CNAM)); Gilbert Laffond (LIRSA-CRC - Laboratoire Interdisciplinaire de Recherche en Sciences de l'Action - Centre de recherche en comptabilité - Conservatoire National des Arts et Métiers (CNAM) : EA4603)
    Abstract: We show that the number of individuals selecting their worst alternatives within a finite set of alternatives can be written as an alternating sum of the number of individuals having their best choice within subset of alternatives. The identities are then applied to random utility models, including the multinomial logit model, the mixed logit model and the disaggregated version of the CES representative consumer model. Finally, we show that better estimates are obtained if respondents are asked to reveal their worst instead of their best choices.
    Keywords: Best-worst; CES; Discrete choice models; Gumbel distribution; Logit; Logsum
    Date: 2013–05–24
  2. By: Francesco Bartolucci (University of Perugia); Federico Belotti (University of Rome "Tor Vergata"); Franco Peracchi (University of Rome "Tor Vergata" and EIEF)
    Abstract: Recent literature on panel data has emphasized the importance of accounting for time-varying unobserved heterogeneity, which may stem either from time-varying omitted variables or macro-level shocks that affect each individual unit differently. In this paper, we propose a computationally convenient test for the null hypothesis of time-invariant individual effects. The proposed test is an application of Hausman (1978) specification test procedure and can be applied to generalized linear models for panel data, a wide class of models that includes the Gaussian linear model and a variety of nonlinear models typically employed for discrete or categorical outcomes. The basic idea is to compare fixed effects estimators defined as the maximand of full and pairwise conditional likelihood functions. Thus, the proposed approach requires no assumptions on the distribution of the individual effects and, most importantly, it does not require them to be independent of the covariates in the model. We investigate the finite sample properties of the test through a set of Monte Carlo experiments. Our results show that the test performs quite well, with small size distortions and good power properties. A health economics example based on data from the Health and Retirement Study is used to illustrate the proposed test.
    Date: 2013
  3. By: Pfarr, Christian; Schmid, Andreas; Ulrich, Volker
    Abstract: More than twenty years after the fall of the iron curtain, do citizens from former Communist countries still exhibit attitudes and preferences with regard to the welfare state and income redistribution that differ from those in the West? This paper seeks to answer this question for Germany after reunification using not only survey data on attitudes but also evidence on preferences from a discrete choice experiment, both based on a representative sample. In a first step, we revisit the empirical evidence, compare our results to those of Alesina and Fuchs-Schündeln (2007) and test whether convergence of attitudes has yet been achieved. In a second step, we apply an advanced method to investigate preferences for redistribution in terms of willingness to pay. This framework is more in line with standard public choice theory as individuals are forced to overcome trade-offs and are exposed to take their inherent budget constraint into account when voicing their opinion on redistribution. The results are quite mixed. While East Germans seem to desire a higher amount of redistribution than West Germans, they are not willing to contribute more through taxation. This finding has important implications for social policy in reunified Germany.
    Keywords: Income redistribution; preferences; willingness to pay;
    JEL: C93 D63 H23 P26
    Date: 2013–05
  4. By: Vitaliy Oryshchenko; Richard J. Smith
    Abstract: If additional information about the distribution of a random variable is available in the form of moment conditions, a weighted kernel density estimate reflecting the extra information can be constructed by replacing the uniform weights with the generalised empirical likelihood probabilities.  It is shown that the resultant density estimator provides an improved approximation to the moment constraints.  Moreover, a reduction in variance is achieved due to the systematic use of the extra moment information.
    Keywords: Weighted kernel density estimation, moment conditions, higher-order expansions, normal mixtures
    JEL: C14
    Date: 2013–03–02
  5. By: Clifton, Judith; Díaz-Fuentes, Daniel; Fernández-Gutiérrez, Marcos
    Abstract: Augmenting consumer welfare was a key justification behind the reform of utilities from the 1980s. But, three decades later, evidence is mounting that consumer satisfaction with household utilities is quite uneven. Moreover, governments, regulators and international organizations are increasingly recognizing that consumers from specific socio-economic backgrounds may be less satisfied than those from other backgrounds. To attend to this, instances of demand-side regulation have been implemented, but there remains a lack of empirical research on the precise links between consumers’ socio-economic background and their satisfaction. This article contrasts consumers’ stated and revealed preferences for three major household utility services (electricity, gas and telecommunications, including internet) across twelve European countries. Contrasting stated and revealed preferences has been applied to policy on transportation, marketing and the environment: this article pioneers the application of this technique to the analysis of satisfaction with household utilities across multiple countries. We find strong evidence that consumers’ socio-economic category matters: consumers with lower levels of education, the elderly and those who are not employed exhibit particular expenditure patterns and lower satisfaction levels vis-à-vis some of or all the services under analysis. We conclude by highlighting how our findings may be of use to regulators in the ongoing quest to improve the quality of utility regulation.
    Keywords: Utilities, regulation, satisfaction, socio-economic background, consumers, stated and revealed preferences.
    JEL: D12 D18 L9 L97 L98
    Date: 2013–05–29
  6. By: Francesco Bartolucci (University of Perugia); Valentino Dardanoni (University of Palermo); Franco Peracchi (University of Rome "Tor Vergata" and EIEF)
    Abstract: We propose a strategy for ranking scientific journals starting from a set of available quantitative indicators that represent imperfect measures of the unobservable "value" of the journals of interest. After discretizing the available indicators, we estimate a latent class model for polytomous item response data and use the estimated model to classify each journal. We apply the proposed approach to data from the Research Evaluation Exercise (VQR) carried out in Italy with reference to the period 2004-2010, focusing on the sub-area consisting of Statistics and Financial Mathematics. Using four quantitative indicators of the journals' scientific value (IF, IF5, AIS, h-index), some of which not available for all journals, we derive a complete ordering of the journals according to their latent value. We show that the proposed methodology is relatively simple to implement, even when the aim is to classify journals into finite ordered groups of a fixed size. Finally, we analyze the robustness of the obtained ranking with respect to different discretization rules.
    Date: 2013
  7. By: Marc Willinger; Mohamed Ali Bchir; Carine Heitz
    Abstract: We study the evolution of preferences of villagers living under the threat of natural hazards in a volcanic area (Mount Mer- api, Central Java). Between December 2010 and April 2011, shortly after a major eruption, the villagers (perceived) expo- sure changed dramatically. We ran incentivized experiments at the beginning and after this period in order to elicit villagers' risk preferences and time preferences. Our three main …ndings are as follows: (1) there exists a signi…cant negative correlation between risk-tolerance and impatience (before and after): individuals who are more risk-tolerant are less impatient; (2) most respondents exhibit a change in their risk preference and/or time preference after having been exposed to a higher level of threat; (3) there ex- ists a signi…cant negative correlation between the erosion of risk tolerance and the erosion of patience: individuals who became less (more) risk-tolerant also became more (less) impatient.
    Date: 2013–05
  8. By: Stephen Bazen (AMSE - Aix-Marseille School of Economics - Aix-Marseille Univ. - Centre national de la recherche scientifique (CNRS) - École des Hautes Études en Sciences Sociales [EHESS] - Ecole Centrale Marseille (ECM)); Xavier Joutard (AMSE - Aix-Marseille School of Economics - Aix-Marseille Univ. - Centre national de la recherche scientifique (CNRS) - École des Hautes Études en Sciences Sociales [EHESS] - Ecole Centrale Marseille (ECM))
    Abstract: The widely used Oaxaca decomposition applies to linear models. Extending it to commonly used nonlinear models such as binary choice and duration models is not straightforward. This paper shows that the original decomposition using a linear model can be obtained as a first order Taylor expansion. This basis provides a means of obtaining a coherent and unified approach which applies to nonlinear models, which we refer to as a Taylor decomposition. Explicit formulae are provided for the Taylor decomposition for the main nonlinear models used in applied econometrics including the Probit binary choice and Weibull duration models. The detailed decomposition of the explained component is expressed in terms of what are usually referred to as marginal effects and a remainder. Given Jensen's inequality, the latter will always be present in nonlinear models unless an ad hoc or tautological basis for decomposition is used.
    Keywords: Oaxaca decomposition; nonlinear models
    Date: 2013–05
  9. By: Cem BaÅŸlevent (Department of Economics, Istanbul Bilgi University); Hasan KirmanoÄŸlu (Department of Economics, Istanbul Bilgi University)
    Abstract: We examine whether employees’ preferences for various job attributes are associated with their individual characteristics in ways that are in line with ‘hierarchy of needs’ theories. Using data from the fifth round of the European Social Survey, we observe the influence of socio-demographic and dispositional characteristics as well as socialization experiences on opinions regarding the importance of five different desirable job attributes. An item-by-item examination of the attributes (including ‘security’ and ‘offering a high income’) reveals that dispositional factors (measured using the battery of items in Schwartz’s theory of basic personal values) influence job attitudes in expected ways, but employees also tend to place more importance on attributes that concern them more directly. For example, while female employees care more about being able to combine work and family responsibilities, younger workers value training opportunities more highly than older ones. Regarding socialization experiences, we find that job security is more important for those who have been unemployed in the past. We interpret our findings to mean that hierarchy of needs theories are valid in the context of job attitudes in the sense that the ranking of preferred job attributes is quite predictable once individual characteristics are accounted for.
    Keywords: preferred job attributes, hierarchy of needs, basic personal values, European Social Survey
    JEL: C25 J28
    Date: 2012–01
  10. By: Xiaohong Chen (Institute for Fiscal Studies and Yale University); Maria Ponomareva; Elie Tamer (Institute for Fiscal Studies and Northwestern University)
    Abstract: Parametric mixture models are commonly used in applied work, especially empirical economics, where these models are often employed to learn for example about the proportions of various types in a given population. This paper examines the inference question on the proportions (mixing probability) in a simply mixture model in the presence of nuisance parameters when sample size is large. It is well known that likelihood inference in mixture models is complicated due to 1) lack of point identification, and 2) parameters (for example, mixing probabilities) whose true value may lie on the boundary of the parameter space. These issues cause the profiled likelihood ration (PLR) statistic to admit asymptotic limits that differ discontinuously depending on how the true density of the data approaches the regions of singularities where there is lack of point identification. This lack of uniformity in the asymptotic distribution suggests that confidence intervals based on pointwise asymptotic approximations might lead to faulty inferences. This paper examines this problem in details in a finite mixture model and provides possible fixes based on the parametric bootstrap. We examine the performance of this parametric bootstrap in Monte Carlo experiments and apply it to data from Beauty Contest experiments. We also examine small sample inferences and projection methods.
    Keywords: Finite mixtures, parametric bootstrap, profiled likelihood ratio statistic, partial identification, parameter on the boundary
    Date: 2013–05

This nep-dcm issue is ©2013 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.