nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2014‒02‒02
seven papers chosen by
Edoardo Marcucci
Universita' di Roma Tre

  1. Demand Modeling, Forecasting, and Counterfactuals, Part I By Parag A. Pathak; Peng Shi
  2. On the Tractability of the Piecewiselinear Approximation for General Discrete-Choice Network Revenue Management By Sumit Kunnumkal; Kalyan Talluri
  3. Empirical welfare analysis with preference heterogeneity By Haan, Peter; Decoster, Andre
  4. Skills and the graduate recruitment process: Evidence from two discrete choice experiments By Velden R.K.W. van der; Humburg M.
  5. Global Sourcing: Towards an Empirical Test of the Hold-up Model By Kohler, Wilhelm; Smolka, Marcel
  6. Adverse selection and heterogeneity of demand responsiveness By Normann Lorenz
  7. Stepping Forward: Personality Traits, Choice of Profession, and the Decision to Become Self-Employed By Sorgner, Alina; Fritsch, Michael

  1. By: Parag A. Pathak; Peng Shi
    Abstract: There are relatively few systematic comparisons of the ex ante counterfactual predictions from structural models to what occurs ex post. This paper uses a large-scale policy change in Boston in 2014 to investigate the performance of discrete choice models of demand compared to simpler alternatives. In 2013, Boston Public Schools (BPS) proposed alternative zone configurations in their school choice plan, each of which alters the set of schools participants are allowed to rank. Pathak and Shi (2013) estimated discrete choice models of demand using families' historical choices and these demand models were used to forecast the outcomes under alternative plans. BPS, the school committee, and the public used these forecasts to compare alternatives and eventually adopt a new plan for Spring 2014. This paper updates the forecasts using the most recently available historical data on participants' submitted preferences and also makes forecasts based on an alternative statistical model not based a random utility foundation. We describe our analysis plan, the methodology, and the target forecast outcomes. Our ex ante forecasts eliminate any scope for post-analysis bias because they are made before new preferences are submitted. Part II will use newly submitted preference data to evaluate these forecasts and assess the strengths and limitations of discrete choice models of demand in our context.
    JEL: H52 I21
    Date: 2014–01
  2. By: Sumit Kunnumkal; Kalyan Talluri
    Abstract: The choice network revenue management (RM) model incorporates customer purchase behavior as customers purchasing products with certain probabilities that are a function of the offered assortment of products, and is the appropriate model for airline and hotel network revenue management, dynamic sales of bundles, and dynamic assortment optimization. The underlying stochastic dynamic program is intractable and even its certainty-equivalence approximation, in the form of a linear program called Choice Deterministic Linear Program (CDLP) is difficult to solve in most cases. The separation problem for CDLP is NP-complete for MNL with just two segments when their consideration sets overlap; the affine approximation of the dynamic program is NP-complete for even a single-segment MNL. This is in contrast to the independent-class (perfect-segmentation) case where even the piecewise-linear approximation has been shown to be tractable. In this paper we investigate the piecewise-linear approximation for network RM under a general discrete-choice model of demand. We show that the gap between the CDLP and the piecewise-linear bounds is within a factor of at most 2. We then show that the piecewise-linear approximation is polynomially-time solvable for a fixed consideration set size, bringing it into the realm of tractability for small consideration sets; small consideration sets are a reasonable modeling tradeoff in many practical applications. Our solution relies on showing that for any discrete-choice model the separation problem for the linear program of the piecewise-linear approximation can be solved exactly by a Lagrangian relaxation. We give modeling extensions and show by numerical experiments the improvements from using piecewise-linear approximation functions.
    JEL: C61 L93 L83 M11
    Date: 2014–01
  3. By: Haan, Peter; Decoster, Andre
    Abstract: We apply recently proposed individual welfare measures in the context of preference heterogeneity, derived from structural labour supply models. Contrary to the standard practice of using reference preferences and wages, these measures preserve preference heterogeneity in the normative step of the analysis. They also make the ethical priors, implicit in any interpersonal comparison, more explicit. Information on preference heterogeneity is obtained from a structural discrete choice labor supply model for married women estimated on microdata from the Socio Economic Panel (SOEP) in Germany. We construct welfare orderings of households according to the different metrics, each embodying different ethical choices concerning the treatment of preference heterogeneity in the consumption-leisure space and provide empirical evidence about the sensitivity of the welfare orderings to different normative principles. We also discuss how sensitive the assessment of a tax reform is to the choice of different metrics. --
    JEL: C35 D63 H24
    Date: 2013
  4. By: Velden R.K.W. van der; Humburg M. (ROA)
    Abstract: In this study we elicit employers preferences for a variety of CV attributes and types of skills when recruiting university graduates. Using two discrete choice experiments, we simulate the two common steps of the graduate recruitment process 1 the selection of suitable candidates for job interviews based on CVs, and 2 the hiring of graduates based on observed skills. We show that in the first step, employers attach most value to CV attributes which signal a high stock of occupation-specific human capital indicating low training costs and short adjustment periods; attributes such as relevant work experience and a good match between the field of study and the job tasks. In line with the preferences in the first step, employers actual hiring decision is mostly influenced by graduates level of professional expertise and interpersonal skills. Other types of skills also play a role in the hiring decision but are less important, and can therefore not easily compensate for a lack of occupation-specific human capital and interpersonal skills.
    Keywords: Analysis of Education; Human Capital; Skills; Occupational Choice; Labor Productivity;
    JEL: J24 I21
    Date: 2014
  5. By: Kohler, Wilhelm; Smolka, Marcel
    Abstract: We use Spanish firm-level data to test the hold-up model of global sourcing proposed by Antr s & Helpman (2004). We propose a novel representation of the model which guides us in bringing the theory to the data. We estimate a discrete choice model of firms' sourcing behavior, separately for the location choice and the ownership choice of sourcing. We find that a firm's productivity interacts with an industry's headquarter intensity in governing both dimensions of sourcing in the way suggested by the Antras-Helpman model. In addition, estimated ``ex ante sourcing premia'' of firms lend brought support to the idea that firms self-select into sourcing activities. --
    JEL: F12 F23 F14
    Date: 2013
  6. By: Normann Lorenz
    Abstract: This paper analyzes the distortions of (health) insurers' benefit levels due to adverse selection if individuals' responsiveness to differences in contracts is heterogeneous. Within a discrete choice model with two risk types and imperfect competition the following results are shown: In the pooling equilibrium, a positive correlation of low risk and high responsiveness (e.g., younger individuals being both healthier and faster to switch insurers than older individuals) increases the distortion of the uniform benefit level if the share of low risks is small; if the share of low risks is large, the reverse holds, but only if the average level of responsiveness is high. In the separating equilibrium, a positive correlation increases the distortion of the contract for the low risks, unless the number of insurers offering the contract for the high risks is very small or a large share of the high risks chooses the contract designated for the low risks. These results imply that the welfare effects of a policy intervention of making individuals more responsive crucially depend on which risk types' responsiveness is increased more. The results also have implications for the estimation of the level of risk aversion and of the welfare effects of adverse selection.
    Keywords: Adverse selection, discrete choice
    JEL: I13
    Date: 2014
  7. By: Sorgner, Alina; Fritsch, Michael
    Abstract: We argue that entrepreneurial choice proceeds in at least two steps, with vocational choice nearly always preceding choice of employment status, whether that be self-employment or dependent employment. Since the two decisions are interrelated, analysis of entrepreneurial choice as a single act may lead to inconsistent estimates of the factors that determine the decision to launch a business venture. Our empirical analysis utilizes a bivariate probit model that jointly estimates both decisions. The results support our argument that entrepreneurial choice is a two-stage decision process. --
    JEL: L26 J24 D01
    Date: 2013

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.