nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2014‒04‒11
ten papers chosen by
Edoardo Marcucci
Universita' di Roma Tre

  1. Household Fuel Choice in Urban Ethiopia: A Random Effects Multinomial Logit Analysis By Alem, Yonas; Beyene, Abebe D.; Kohlin, Gunnar; Mekonnen, Alemu
  2. The Economics of Presenteeism: A discrete choice & count model framework By Pedersen, Kjeld Møller; Skagen, Kristian
  3. Effect of an Accessibility Measure in a Model for Choice of Residential Location, Workplace, and Type of Employment By Ignacio Inoa; Nathalie Picard; André De Palma
  4. Intra-household Decision Models of Residential and Job Location By Nathalie Picard; André De Palma; Ignacio Inoa
  5. Behavioral and Descriptive Forms of Choice Models By Ariel Pakes
  6. A discrete choice model approximation to the consumer’s choice among television displays By Carlos Giovanni González Espitia; Natalia Serna Borrero
  7. Misperception of Consumption: Evidence from a Choice Experiment By Seeun Jung; Yasuhiro Nakamoto; Masayuki Sato; Katsunori Yamada
  8. Are taxes beautiful? A survey experiment on information, tax choice and perceived adequacy of the tax burden By Abbiati, Lorenzo; Antinyan, Armenak; Corazzini, Luca
  9. A Smooth Transition Logit Model of the Effects of Deregulation in the Electricity Market By A.S. Hurn; Annastiina Silvennoinen; Timo Teräsvirta
  10. Bidimensional Matching with Heterogeneous Preferences: Smoking in the Marriage Market By Climent Quintana-Domeque; Pierre-Andre Chiappori; Sonia Oreffice

  1. By: Alem, Yonas; Beyene, Abebe D.; Kohlin, Gunnar; Mekonnen, Alemu
    Abstract: We use three rounds of a rich panel data set to investigate the determinants of household fuel choice and energy transition in urban Ethiopia. We observe that energy transition did not occur following economic growth in Ethiopia during the past decade. Regression results from a random effects multinomial logit model, which controls for unobserved household heterogeneity, show that households’ economic status, price of alternative energy sources, and education are important determinants of fuel choice in urban Ethiopia. The results also suggest the use of multiple fuels, or “fuel stacking behavior.” We argue that policy makers could target these variables to encourage transition to cleaner energy sources.
    Keywords: urban Ethiopia, energy choice, random effects multinomial logit
    JEL: C25 Q23 Q42 O13
    Date: 2013–10–10
  2. By: Pedersen, Kjeld Møller (COHERE, Department of Business and Economics); Skagen, Kristian (COHERE, Department of Business and Economics)
    Abstract: There are three levels in this paper: A search for economic theories about presenteeism, a search for appropriate econometric approaches, and finally empirical results based on a unique Danish cross sectional data set. There are two economic approaches to presenteeism: 1. Productivity losses and 2. labor supply. The first is part of the indirect cost component in cost-of-illness studies and economic evaluation. There are two core questions in the productivity loss literature: Measurement of productivity losses (‘how much’) which has dominated the research agenda and valuation of incurred productivity losses (monetary value). Few economists have addressed the valuation issue and point out that the wage rate sometimes is inadequate. The starting point in the labor supply literature is sickness absence coupled with labor demand. The few economic models about presenteeism are explored and found lacking in the sense that they do not capture the essence of presenteeism. However, discrete choice models (random utility models) seem to be adequate in that the choice about going sick to work basically is a discrete choice situation that can be extended to include discrete counts, i.e. episodes of presenteeism within a given time period. The econometrics of presenteeism must have count models as the starting point due to the many zeroes, i.e. many persons do not experience presenteeism and, if they do, usually relatively few days (‘events’) in a given period and the discrete choice nature of presenteeism. Drawing on the econometric literature on utilization of medical services, the following models are discussed briefly: Poisson models, negative binominal, zero-inflated negative binomial, two part models (hurdle models) and latent class models (finite mixture models). This is in contrast to almost all previous literature where logistic regression has been the dominant statistical strategy. The Poissson model is discarded because an important feature (mean – variance) does not hold. The other models are all used in the empirical part of the paper, and an attempt at model selection is made. The empirical analyses are based on a cross-section survey of Danes in the labor force, N=4,060. The survey was designed with presenteeism in mind – one of the few available data sets at present. Ideally, theory/models should guide empirical work, but can do so only if fully specified theories are available and this is not the case for the random utility models that do no provide much guidance on relevant explanatory variables. The explanatory variables therefore are selected from the existing empirical works along with a number of new variables used in the survey, e.g. attitudinal variables about presenteeism and sickness absence and questions about work environment. A consistent result across all analyses is – not surprisingly - the importance of self reported health status: The worse health situation, the more presenteeism. . Another consistent result is that sickness absence and presenteeism are positively correlated. Persons with managerial positions also consistently have more presenteeism Age and genders are also (almost) consistently statistically significant. Fear of unemployment is also consistently and significantly related to presenteeism.
    Keywords: Presenteeism; sickness absence; labor supply; cost-of-illness; economic evaluation; count models
    JEL: C35 I12 J22 J24
    Date: 2014–02–15
  3. By: Ignacio Inoa (THEMA - Théorie économique, modélisation et applications - CNRS : UMR8184 - Université de Cergy Pontoise); Nathalie Picard (THEMA - Théorie économique, modélisation et applications - CNRS : UMR8184 - Université de Cergy Pontoise, Department of Economics, Ecole Polytechnique - CNRS : UMR7176 - Polytechnique - X); André De Palma (Department of Economics, Ecole Polytechnique - CNRS : UMR7176 - Polytechnique - X, ENS Cachan - École Normale Supérieure de Cachan - École normale supérieure [ENS] - Cachan)
    Abstract: A three-level nested logit model for the choice of residential location, workplace, and type of employment is used to assess the effect of an individual-specific measure of accessibility to employments that takes into account the attractiveness of different occupations when the choice of workplace is anticipated in the decision of residential location. The model allows for variation in the preferences for types of employment across individuals and accounts for individual heterogeneity of preferences at each choice level in education, age, gender, and children. Using data from the Île-de-France region, it shows that the individual specific accessibility measure is an important determinant of the choice of residential location and its effect differs along the life cycle. The attractiveness of the types of employment is a better predictor of the workplace location than the usual total number of employments.
    Keywords: residential location, employment location, accessibility, nested logit, Île-de-France.
    Date: 2014–03–24
  4. By: Nathalie Picard (THEMA - Théorie économique, modélisation et applications - CNRS : UMR8184 - Université de Cergy Pontoise, Department of Economics, Ecole Polytechnique - CNRS : UMR7176 - Polytechnique - X); André De Palma (Department of Economics, Ecole Polytechnique - CNRS : UMR7176 - Polytechnique - X, ENS Cachan - École Normale Supérieure de Cachan - École normale supérieure [ENS] - Cachan); Ignacio Inoa (THEMA - Théorie économique, modélisation et applications - CNRS : UMR8184 - Université de Cergy Pontoise)
    Abstract: Residential location decision is often a household joint decision involving several decision-makers. These different decision-makers usually have diverging preferences, especially in dual-earner households, when spouses work at different locations. Since about half a century, literature on residential location has studied in great detail the influence of socio-demographic characteristics (and in particular the differences between females and males or between multiple-worker and single-worker households). However, there is no research devoted to the within-family joint decision process leading to residential location decision (and work-place decisions). In the context of Paris Area, we analyze differences between spouses' values of commuting times and show that spouses' disparities in commuting decisions is a key element in the intra-household decision process. The single-worker household approach leaves aside by construction important intra-household considerations that influence commuting time and accessibility to jobs. We review different models useful to study intra-household decisions in dual-earner households. To do that, we base our analysis on the framework introduced by Chiappori, de Palma, Picard, and Inoa (2013), which applies the collective approach of household behavior(Chiappori, 1988; Chiappori, 1992) to describe residential location choice of dual-earner households. This collective approach has been used in several economic fields, but not in urban and transport economics so far. Furthermore, we argue that the framework developed by Inoa, Picard, and de Palma (2013), can also be adapted to analyze the joint residential and job location decisions in a two-worker household. The analysis is based on two accessibility variables (one for each spouse) embedded in a three-level nested Logit model which is used to study the interdependence of residential and workplace locations, while accounting for variation of preferences for job types across individuals.
    Keywords: intra-household interaction, residential location, Paris region
    Date: 2014–03–24
  5. By: Ariel Pakes
    Abstract: Empirical work on choice models, especially work on relatively new topics or data sets, often starts with descriptive, or what is often colloquially referred to as "reduced form", results. Our descriptive form formalizes this process. It is derived from the underlying behavioral model, has an interpretation in terms of fit, and can sometimes be used to quantify biases in agents' expectations. We consider estimators for the descriptive form of discrete choice models with (and without) interacting agents that take account of approximation errors as well as unobservable sources of endogeneity. We conclude with an investigation of the descriptive form of two period entry models.
    JEL: B4 C51
    Date: 2014–03
  6. By: Carlos Giovanni González Espitia (Departamento de Economía. Universidad ICESI.); Natalia Serna Borrero (Banco de la República, sucursal Cali.)
    Abstract: The consumer’s choice over a bundle of products depends on the observable and unobservable characteristics of the product and the consumer. The choice is made over the basis of maximizing utility subject to their income restrictions and, at the same time, firms make product differentiation decisions over the basis of maximizing profit. Quality is one way to differentiate products. An example of this type of differentiation happens in the TV market where several displays are developed. Our objective is to determine the probability for a consumer of choosing a type of display among five kinds: standard tube, LCD, plasma, projection and LED. Using a multinomial logit, we find that electronic appliances such as DVDs and audio systems, as well as socioeconomic status, increase the probability of choosing a high-tech display television. Our empirical approximation contributes to the further understanding of consumer rational behavior through the theory utility maximization and highlights the importance of studying the market structure and analyzing changes in welfare and efficiency.
    Keywords: Observable product characteristics, product differentiation, quality, television display
    JEL: D12 L15
    Date: 2014
  7. By: Seeun Jung (PSE - Paris-Jourdan Sciences Economiques - CNRS : UMR8545 - École des Hautes Études en Sciences Sociales (EHESS) - École des Ponts ParisTech (ENPC) - École normale supérieure [ENS] - Paris - Institut national de la recherche agronomique (INRA), EEP-PSE - Ecole d'Économie de Paris - Paris School of Economics - Ecole d'Économie de Paris); Yasuhiro Nakamoto (Kyushu Sangyo University - Kyushu Sangyo University); Masayuki Sato (Graduate School of Human Development and Environment - Kobe University); Katsunori Yamada (ISER - Institute of Social and Economic Research - Osaka University)
    Abstract: We investigate people's different conceptions of the economic term "consumption" when comparing with others. An Internet-based hypothetical discrete choice experiment was conducted with Japanese participants. As in other relative income comparison studies, we found that own consumption and own saving had a positive impact on utility, whereas the consumption and saving of a reference person had a negative impact on utility. However, the results show that the magnitudes of consumption and saving differ in size; saving could affect utility much more than consumption for the Japanese subjects. By using scope tests, we found that the impact of own consumption is not monotonic and so does not necessarily increase utility. This calls into question the conventional assumption of the monotonicity of "the utility of consumption"; consumption could be perceived as a negative good. Our results, therefore, provide some evidence that, in reality, people understand and perceive the economic terms differently from what economists would expect. Furthermore, when considering the consumption of others as well as their own, the size of the discrepancy is even bigger.
    Keywords: Relative Utility ; Choice Experiment ; Misperception of Economic Terms
    Date: 2014–03
  8. By: Abbiati, Lorenzo; Antinyan, Armenak; Corazzini, Luca
    Abstract: We report results from a survey experiment aimed at testing whether providing information on the national public expenditure to the taxpayers and whether involving them in the process of allocating tax revenues over public goods influence the level of the adequate tax rate - the fraction of income that individuals consider adequate to pay as taxes. We find that providing information on public expenditure does not influence the level of the adequate tax rate. On the contrary, the level of the adequate tax rate substantially increases when taxpayers can get to choose the public goods to finance through their taxation. --
    Keywords: Tax Choice,Adequate Tax Rate,Survey Experiment
    JEL: H24 H50 D31
    Date: 2014
  9. By: A.S. Hurn (School of Economics and Finance, Queensland University of Technology); Annastiina Silvennoinen (School of Economics and Finance, Queensland University of Technology); Timo Teräsvirta (Aarhus University and CREATES)
    Abstract: We consider a nonlinear vector model called the logistic vector smooth transition autoregressive model. The bivariate single-transition vector smooth transition regression model of Camacho (2004) is generalised to a multivariate and multitransition one. A modelling strategy consisting of specification, including testing linearity, estimation and evaluation of these models is constructed. Nonlinear least squares estimation of the parameters of the model is discussed. Evaluation by misspecification tests is carried out using tests derived in a companion paper. The use of the modelling strategy is illustrated by two applications. In the first one, the dynamic relationship between the US gasoline price and consumption is studied and possible asymmetries in it considered. The second application consists of modelling two well known Icelandic riverflow series, previously considered by many hydrologists and time series analysts. JEL Classification: C23, C51, L94, Q41.
    Keywords: Smooth transition, binary choice model, logit model, electricity spot prices, peak load pricing, price spikes
    Date: 2014–03–28
  10. By: Climent Quintana-Domeque; Pierre-Andre Chiappori; Sonia Oreffice
    Abstract: We develop a bidimensional matching model under transferable utility, where individuals are characterized by a continuous trait (e.g., socioeconomic status) and a binary attribute (e.g., smoking status).� The model is "truly multidimensional", in the sense that the impact of the traits cannot be summarized by a one-dimensional index.� We present a general resolution strategy based on optimal control theory, and characterize the stable matching.� We derive testable predictions about equilibrium matching patterns.� Using US data, we find that the observed marital sorting of smokers and non-smokers by education is consistent with our model.
    Keywords: Marriage market, multidimensional matching, continuous and discrete characteristics, heterogeneous preferences
    JEL: D1 J1
    Date: 2014–01–29

This nep-dcm issue is ©2014 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.