nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2006‒11‒25
nine papers chosen by
Philip Yu
Hong Kong University

  1. Lexicographic Preferences in Discrete Choice Experiments: Consequences on Individual-Specific Willingness to Pay Estimates By Danny Campbell; W. George Hutchinson; Riccardo Scarpa
  2. Can Risk Aversion Explain Schooling Attainments? Evidence From Italy By Christian Belzil; Marco Leonardi
  3. Self-selection in migration and returns to unobservable skills By Benoit Dostie; Pierre Thomas Léger
  4. Workfare: Eine wirksame Alternative zum Kombilohn By Holger Bonin; Hilmar Schneider
  5. Choice of new attributes in the "Elimination by Aspects" duopoly. By Reynald-Alexandre Laurent
  6. Public Pension Programmes and the Retirement of Married Couples in Denmark By Paul Bingley; Gauthier Lanot
  7. On the choice of most-preferred alternatives By Kukushkin, Nikolai S.
  8. Estimation of an Occupational Choice Model when Occupations are Misclassified By Sullivan, Paul
  9. Interpolating Value Functions in Discrete Choice Dynamic Programming Models By Sullivan, Paul

  1. By: Danny Campbell (Gibson Institute of Land, Food and Environmen, Queen’s University Belfast); W. George Hutchinson (Gibson Institute of Land, Food and Environmen, Queen’s University Belfast); Riccardo Scarpa (University of Waikato)
    Abstract: In discrete choice experiments respondents are generally assumed to consider all of the attributes across each of the alternatives, and to choose their most preferred. However, results in this paper indicate that many respondents employ simplified lexicographic decision-making rules, whereby they have a ranking of the attributes, but their choice of an alternative is based solely on the level of their most important attribute(s). Not accounting for these simple decision-making heuristics introduces systemic errors and leads to biased point estimates, as they are a violation of the continuity axiom and a departure from the use of compensatory decision-making. In this paper the implications of lexicographic preferences are examined. In particular, using a mixed logit specification this paper investigates the sensitivity of individual-specific willingness to pay (WTP) estimates conditional on whether lexicographic decision-making rules are accounted for in the modelling of discrete choice responses. Empirical results are obtained from a discrete choice experiment that was carried out to address the value of a number of rural landscape attributes in Ireland.
    Keywords: Continuity axiom, Discrete Choice Experiments, Lexicographic Preferences, Mixed Logit, Individual-Specific Willingness to Pay
    JEL: C35 Q24 Q51
    Date: 2006–10
  2. By: Christian Belzil (GATE CNRS); Marco Leonardi (Università di Milano)
    Abstract: Using unique Italian panel data, in which individual differences in behavior toward risk are measured from answers to a lottery question, we investigate if (and to what extent) risk aversion can explain differences in schooling attainments. We formulate the schooling decision process as a reduced-form dynamic discrete choice. The model is estimated with a degree of flexibility virtually compatible with semiparametric likelihood techniques. We analyze how grade transition from one level to the next varies with preference heterogeneity (risk aversion), parental human capital, socioeconomic variables and persistent unobserved (to the econometrician) heterogeneity. We present evidence that schooling attainments decrease with risk aversion, but despite a statistically significant effect, differences in attitudes toward risk account for a modest portion of the probability of entering higher education. Differences in ability(ies) and in parental human capital are much more important. in the most general version of the model, the likelihood function is the joint probability of schooling attainments, and post-schooling wealth and risk aversion.
    Keywords: dynamic discrete choices, éducation, human capital, risk aversion
    JEL: J24
    Date: 2006–10
  3. By: Benoit Dostie (IEA, HEC Montréal); Pierre Thomas Léger (IEA, HEC Montréal)
    Abstract: Several papers have tested the empirical validity of the migration models proposed by Borjas (1987) and Borjas, Bronars, and Trejo (1992). However, to our knowledge, none has been able to disentangle the separate impact of observable and unobservable individual characteristics, and their respective returns across different locations, on an individual's decision to migrate. We build a model in which individuals sort, in part, on potential earnings - where earnings across different locations are a function of both observable and unobservable characteristics. We focus on the inter-provincial migration patterns of Canadian physicians. We choose this particular group for several reasons including the fact that they are paid on a fee-for-service basis. Since wage rates are exogenous, earning differentials are driven by differences in productivity. We then estimate a mixed conditional-logit model to determine the effects of individual and destination-specific characteristics (particularly earnings differentials) on physician location decisions. We find, among other things, that high-productivity physicians (based on unobservables) are more likely to migrate to provinces where the productivity premium is greater, while low-productivity physicians are more likely to migrate to areas where the productivity premium is lower. These results are consistent with a modified Borjas model of self-selection in migration based on both unobservables and observables.
    Keywords: Migration, self-selection, earnings, longitudinal data, productivity.
    JEL: J24 J61 C23 C35
    Date: 2006–01
  4. By: Holger Bonin (IZA Bonn); Hilmar Schneider (IZA Bonn and DIW Berlin)
    Abstract: In the debate on in-work benefits in Germany it is often overlooked that such subsidies may only be effective if basic minimum income is remarkably decreased for those who are employable. However, proposals following this principle will hardly achieve political consensus as can be derived from reactions to the model of ifo or the board of economic advisors. IZA is therefore proposing workfare as an effective alternative, which may accomplish a strong incentive for the acceptance of low paid jobs without having to cut the current welfare level. Simulations based on a micro-econometric labor supply model show that this could add to the workforce an additional number of 800,000 workers. Roughly the same effect could be achieved by the ifo model, however, at the expense of massive cuts of disposable income for welfare recipients. Hence, workfare turns out as an efficient alternative. Moreover, there is no useful combination between in-work benefits based on the current welfare level and workfare.
    Keywords: in-work benefits, low-wage subsidies, labor supply, discrete choice, Germany
    JEL: J68 J38 H24 J22
    Date: 2006–10
  5. By: Reynald-Alexandre Laurent
    Abstract: The "Elimination by aspects" (EBA) duopoly of product differentiation (Laurent, 2006a) was constructed from the discrete model of probabilistic choice worked out by Tversky (1972a,b). In this framework, an unique price equilibrium exists with a "differentiation by attributes", which embodies horizontal and vertical differentiations as possible special cases. This paper extends this analysis by studying a two-stage game in which firms choose the specific attributes of their product and then compete in prices. At the price equilibrium, the "competitive effect", present in pure vertical differentiation models, is replaced by a "differentiation effect" in this EBA duopoly. Subgame perfect Nash equilibria are shown to exist with exogenous costs but also with attributes-dependent unit and fixed costs. At the equilibrium, products are generally differentiated both horizontally and vertically. But a purely vertical outcome may also occur when costs of innovation are strongly convex or when consumers are very sensible to the price levels. When costs are endogenous, the social optimum is achieved for a pure horizontal differentiation. Thus, there is too much differentiation at the equilibrium: the vertical dimension induces a strong raise of prices, which also reduces the welfare.
    Date: 2006
  6. By: Paul Bingley (Economics Department, Aarhus School of Business); Gauthier Lanot (Keele University, Centre for Economic Research and School of Economic and Management Studies)
    Abstract: In this paper we study the economic determinants of the joint retirement process of married couples. We propose a tractable dynamic discrete choice model for retirement decisions which allows for non-trivial saving behaviour. We estimate the model on a 1\% sample of Danish couples of potential retirement age drawn from a population-based administrative register. The introduction and subsequent reforms of a publicly financed early-retirement programme provide us with variation in the data to ensure identification of the the elasticities of participation/retirement with respect to income flows. Our estimates imply a significant asymmetry in the sensitivity of retirement behaviour of men and women with respect to variation in their own, or their spouse’s income flows.
    Keywords: Retirement, pensions, dynamic structural model.
    JEL: J26 H24 C35
    Date: 2006–10
  7. By: Kukushkin, Nikolai S.
    Abstract: Maximal elements of a binary relation on compact subsets of a metric space define a choice function. Necessary and sufficient conditions are found for: (1) the choice function to have nonempty values and be path independent; (2) the choice function to have nonempty values provided the underlying relation is an interval order. For interval orders and semiorders, the same properties are characterized in terms of representations in a chain.
    Keywords: Maximal element; Path independence; Interval order; Semiorder
    JEL: D71
    Date: 2006–11–09
  8. By: Sullivan, Paul
    Abstract: This paper examines occupational choices using a discrete choice model that accounts for the fact that self-reported occupation data is measured with error. Despite evidence from validation studies which suggests that there is a substantial amount of measurement error in self-reported occupations, existing research has not corrected for classification error when estimating models of occupational choice. This paper develops a panel data model of occupational choices that corrects for misclassification in occupational choices and measurement error in occupation-specific work experience variables. The model is used to estimate the extent of measurement error in self-reported occupation data and quantify the bias that results from ignoring measurement error in occupation codes when studying the determinants of occupational choices and estimating the effects of occupation-specific human capital on wages. The parameter estimates reveal that 9% of occupational choices in the 1979 cohort of the National Longitudinal Survey of Youth are misclassified. Ignoring misclassification biases the median parameter in the occupational choice model by 25%.
    Keywords: occupational choice; misclassification; simulation methods
    JEL: C25 J24 C15
    Date: 2006–11
  9. By: Sullivan, Paul
    Abstract: Structural discrete choice dynamic programming models have been shown to be a valuable tool for analyzing a wide range of economic behavior. A major limitation on the complexity and applicability of these models is the computational burden associated with computing the high dimensional integrals that typically characterize an agent's decision rules. This paper develops a regression based approach to interpolating value functions during the solution of dynamic programming models that alleviates this burden. This approach is suitable for use in models that incorporate unobserved state variables that are serially correlated across time and correlated across choices within a time period. The key assumption is that one unobserved state variable, or error term, in the model is distributed extreme value. Additional error terms that allow for correlation between unobservables across time or across choices within a given time period may be freely incorporated in the model. Value functions are simulated at a fraction of the state space and interpolated at the remaining points using a new regression function based on the extreme value closed form solution for the expected maxima of the value function. This regression function is well suited for use in models with large choice sets and complicated error structures. The performance of the interpolation method appears to be excellent, and it greatly reduces the computational burden of estimating the parameters of a dynamic programming model.
    Keywords: dynamic programming models; interpolation; simulation methods; estimation of dynamic programming models
    JEL: C50 C15
    Date: 2006–10

This nep-dcm issue is ©2006 by Philip Yu. 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.