nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2013‒09‒26
six papers chosen by
Edoardo Marcucci
Universita' di Roma Tre

  1. An empirical study of predicting car type choice in Sweden using cross-validation and feature-selection By Habibi, Shiva; Sundberg, Marcus; Karlström, Anders
  3. Mixed logit modeling in Stata--an overview By Arne Risa Hole
  4. The effect of job insecurity on labour supply By Jara Tamayo, Holguer Xavier
  5. Immigration, Wages, and Education: A Labor Market Equilibrium Structural Model By Joan Llull
  6. Technology Adoption and Energy Efficiency in Irrigation: First Results from a Coordination Game in Andhra Pradesh, India By Müller, Malte; Rommel, Jens

  1. By: Habibi, Shiva (KTH); Sundberg, Marcus (KTH); Karlström, Anders (KTH)
    Abstract: In this paper we analyze the prediction problem and focus on building a multinomial logit model (MNL) to predict accurately, the market shares of new cars in the Swedish car fleet in the short-term future. Also, we investigate whether or not different prediction questions lead to different 'best' models specifications. Most of the studies in the field, take an inference-driven approach to select best models to estimate relevant parameters and project the results to the future, whereas we do take a prediction-driven approach. We use feature (variable) selection and cross-validation algorithms to improve predictive performance of models. These methods have been extensively used in other fields such as marketing but are scarce studies employing them in the choice modeling field. Additionally, we introduce four different prediction questions or loss-functions: overall prediction (log-likelihood), brand market share, ethanol (E85)/brand market share, and total share of ethanol cars and the predicted results of these models are compared. The results show that 'best' models prediction depend different prediction questions to answer. Also, they indicate that log-likelihood does not perform accurately when the objective is to predict a sub-section of population such as total share of E85 cars.
    Keywords: Hold-out sample; Out of sample prediction; Feature selection; Cross validation; Model selection; Car type choice; Discrete choice modeling; Clean vehicles
    JEL: R41
    Date: 2013–09–16
  2. By: Edoardo Marcucci (Università degli Studi di Roma Tre)
    Abstract: This paper reviews a set of articles, based on stated preferences techniques, focusing on logistics managers' preferences for freight transport attributes with the intent of assessing the quality of knowledge applied research has produced, its reliability and transferability. Transport choices are relevant in a globalized economy where aggressive marketing strategies are often used to acquire a competitive advantage in the market. Freight transport modeling has been derived out of the classical four step approach developed for passenger transport. A new behavioral approach is progressively gaining popularity thanks to the specific behavioral focus it is based upon. The review performed indicates that there are some evident shortcomings in the way research has been performed so far but, at the same time, there is also a high potential if corrective actions are taken. In particular substantial improvements could be obtained by: 1) clearly defining research and reporting protocols; 2) define and circumscribe who has to be interviewed in the different freight contexts studied; 3) reporting the contractual relationships governing freight movements; 4) reporting freight details (e.g. volume, value, weight); 5) motivating the attribute selection method used. In summary, there is a need for systematizing the procedures and reporting adopted in applied research by introducing a much higher level of detail and rigor in both defying the object of measurement as well in the experimental design protocols employed.
    Keywords: Transport demand, Stated Preferences, Logistics.
    Date: 2013
  3. By: Arne Risa Hole (University of Sheffield)
    Abstract: The "workhorse" model for analysing discrete choice data, the conditional logit model, can be implemented in Stata using the official clogit and asclogit commands. While widely used, this model has several well-known limitations that have led researchers in various disciplines to consider more flexible alternatives. The mixed logit model extends the standard conditional logit model by allowing one or more of the parameters in the model to be randomly distributed. When one models the choices of individuals (as is common in several disciplines, including economics, marketing, and transport), this allows for preference of heterogeneity among respondents. Other advantages of the mixed logit model include the ability to allow for correlations across observations in cases where an individual made more than one choice, and relaxing the restrictive independence from the irrelevant alternatives property of the conditional logit model. There are a range of commands that can be used to estimate mixed logit models in Stata. With the exception of xtmelogit, the official Stata command for estimating binary mixed logit models, all of them are userwritten. The module that is probably best known is gllamm, but while very flexible, it can be slow when the model includes several random parameters. This talk will focus on alternative commands for estimating logit models, with focus on the mixlogit module. We will also look at alternatives and extensions to mixlogit, including the recent lclogit, bayesmlogit, and gmnl commands. The talk will review the theory behind the methods implemented by these commands and present examples of their use.
    Date: 2013–09–16
  4. By: Jara Tamayo, Holguer Xavier
    Abstract: The aim of this paper is to analyse the efect of job insecurity on labour supply. We propose an extension of traditional discrete choice models of labour supply in order to allow for the introduction of non-pecuniary job attributes in the analysis. In our extended model, the choice alternatives are characterised by bundles of income, hours of work and job insecurity. We compare the predictive power and labour supply elasticities obtained with our model to those of a traditional model where only income and discrete hours choices characterise a job. The results show that once job insecurity is included in the discrete choice alternatives, the predictive power of the model improves significantly. Labour supply elasticities are significantly higher than those obtained with a traditional model and increase with the level of job insecurity. Finally, a decrease of job insecurity at work has a positive and significant effect on participation. Policies aimed at improving working conditions could, in this sense, be useful to create incentives in labour market.
    Date: 2013–09–16
  5. By: Joan Llull
    Abstract: This paper analyzes the effect of immigration on wages taking into account human capital and labor supply adjustments. Using U.S. micro-data for 1967-2007, I estimate a labor market equilibrium model that includes endogenous decisions on education, participation, and occupation, and allows for skill-biased technical change. Results suggest important labor market adjustments that mitigate the effect of immigration on wages. These adjustments include career switches, labor market detachment and changes in schooling decisions, and are heterogeneous across the workforce. The adjustments generate substantial self-selection biases at the lower tail of the wage distribution that are corrected by the estimated model.
    Keywords: immigration, wages, human capital, labor supply, dynamic discrete choice, labor market equilibrium
    JEL: J2 J31 J61
    Date: 2013–09
  6. By: Müller, Malte; Rommel, Jens
    Abstract: Farmers’ technology adoption in electric irrigation has recently been analyzed as a coordination problem. To study how the Pareto-inferior equilibrium, farmers are trapped in, can be overcome we have developed a framed field experiment. Leadership and group size are varied in a full factorial experimental design. Initial results show only minor treatment effects. Further analysis is necessary to account for socio-demographic heterogeneity.
    Keywords: Coordination Game, Energy Efficiency, India, Irrigation, Technology Adoption, Community/Rural/Urban Development, Institutional and Behavioral Economics, Research Methods/ Statistical Methods,
    Date: 2013

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.