nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2005‒06‒05
six papers chosen by
Philip Yu
Hong Kong University

  1. Modeling ‘No-choice’ Responses in Attribute Based Valuation Surveys By Arianne T. de Blaeij; Paulo A.L.D. Nunes; Jeroen C.J.M. van den Bergh
  2. Identification of Options and Policy Instruments for the Internalisation of External Costs of Electricity Generation. Dissemination of External Costs of Electricity Supply Making Electricity External Costs Known to Policy-Makers MAXIMA By Anil Markandya; Alberto Longo
  3. Preferences regarding road transports of hazardous materials using choice experiments - any sign of biases? By Hiselius, Lena Winslott
  4. The Case for Utilitarian Voting By Hillinger, Claude
  5. A comparison between alternative models for environmental ordinal data: Nonlinear PCA vs Rasch Analysis By Pieralda FERRARI; Paola ANNONI; Silvia SALINI
  6. Extending the Merton Model: A Hybrid Approach to Assessing Credit Quality By Alexxandros Benos; George Papanastasopoulos

  1. By: Arianne T. de Blaeij (Centre Utrecht (URU), Utrecht University); Paulo A.L.D. Nunes (Free University); Jeroen C.J.M. van den Bergh (Free University)
    Abstract: We examine the impact of providing a ‘no-choice’ option in an attribute based valuation experiment. The aim of the experiment was to assess monetary values of cockle fishery management practices in the Dutch Wadden Sea for different stakeholder groups, namely Dutch citizens, local residents, and tourists. The current policy debate about the management of the Wadden Sea stresses the fact that individual preferences with respect to cockle-fishery differ. The aim of this paper is to analyze the individual preferences in an objective way. Special attention is given to the influence of including a ‘no-choice option’, which is analyzed using a nested logit model. We test whether the full set of policy options can be considered as close substitutes. The estimation results show that the influence of including the no choice option differs among the stakeholders considered.
    Keywords: Stakeholder valuation, Stated choice method, Multinomial logit model, Binary logit model
    JEL: C25 C29 Q22 Q25 Q51
    Date: 2005–05
  2. By: Anil Markandya (The World Bank); Alberto Longo (University of Bath)
    Abstract: In the present paper, after reviewing the results of the ExternE project and its follow-up stages in the estimation of the external costs of electricity production, we look at the policy instruments for the internalisation of such costs. Emphasis is given to subsidies, such as feed-in tariffs, competitive bidding processes and tradable green certificates to stimulate the use of renewables in the production of electricity. When policy-makers are asked to choose the instrument(s) to internalise the externalities in the electricity production, they have to find a solution that gives the best outcome in terms of efficiency, cost minimisation, impact on the job market, security of energy supply, equity of the instrument, technological innovation, certainty of the level of the internalisation, and feasibility. The choice of the instrument will require some trade-offs among these criteria. Conjoint choice analysis can help in investigating how stakeholders and policy makers trade off the criteria when choosing a policy for the internalisation of the externalities. In this paper we present the first results of a questionnaire that employs conjoint choice questions to find out how policy makers and stakeholders of the electricity market trade off some socio-economic aspects in the selection of the policy instruments for the internalisation of the externalities. The results of this first set of interviews will be useful for further research.
    Keywords: Policy instruments, ExternE, External costs, Electricity, Conjoint choice analysis
    JEL: Q42 Q48 Q51
    Date: 2005–05
  3. By: Hiselius, Lena Winslott (Department of Economics, Lund University)
    Abstract: This paper uses the choice experiment approach to assess people's preferences regarding road transports of hazardous materials. In a mail survey, carried out in Stockholm, the capital of Sweden, changes in exposure to hazardous materials are used as a proxy for changes in accident risk. The results are analysed in the light of an earlier study on transports of hazardous materials by rail. Special attention is given to biases associated with the choice experiment method. The presence of hypothetical bias is studied by the use of self-reported degree of confidence that the respondent would vote the same way in a real referendum. The presence of a focusing effect is studied by an inclusion of information on other fatal risks. The indication is that there are no major differences in individual preferences for hazmat transported by rail or road. The estimates are also dependent on the confidence of stated choices and interpreting this dependence as a hypothetical bias, suggest that this type of bias tends to push estimated values downwards. The findings show that individual background data regarding transports of hazardous materials affect individuals in expected ways and there is no focusing effect.
    Keywords: Biases; Hazardous materials; Risk; Choice experiments
    JEL: C25 D61 D81 R41
    Date: 2005–05–23
  4. By: Hillinger, Claude
    Abstract: Utilitarian voting (UV) is defined in this paper as any voting rule that allows the voter to rank all of the alternatives by means of the scores permitted under a given voting scale. Specific UV rules that have been proposed are approval voting, allowing the scores 0, 1; range voting, allowing all numbers in an interval as scores; evaluative voting, allowing the scores -1, 0, 1. The paper deals extensively with Arrow's impossibility theorem that has been interpreted as precluding a satisfactory voting mechanism. I challenge the relevance of the ordinal framework in which that theorem is expressed and argue that instead utilitarian, i.e. cardinal social choice theory is relevant for voting. I show that justifications of both utilitarian social choice and of majority rule can be modified to derive UV. The most elementary derivation of UV is based on the view that no justification exists for restricting voters? freedom to rank the alternatives on a given scale.
    JEL: D72 D71
    Date: 2005–05
  5. By: Pieralda FERRARI; Paola ANNONI; Silvia SALINI
    Abstract: Two different methodologies, Nonlinear PCA and Partial Credit ver sion of Rasch model, are applied to real data to obtain a quantit ative measure of a latent factor. Specifically the goal is to ide ntify the level of damage of a set of particurarly valuable publi c buildings. Several ordinal variables are observed on each build ing, which describe various aspects of damage severity. Scores o n each bulding are then assigned on the basis of computed 'vulner ability' level.
    Keywords: Nonlinear PCA; Partial Credit Rasch Model; Ordinal Variables
  6. By: Alexxandros Benos (University of Piraeus); George Papanastasopoulos (University of Peloponnese)
    Abstract: In this paper we have combined fundamental analysis and contingent claim analysis into a hybrid model of credit risk measurement. We have extended the standard Merton approach to estimate a new risk-neutral distance to default metric, assuming a more complex capital structure, adjusting for dividend payments, introducing randomness to the default point and allowing a fractional recovery when default occurs. Then, using financial ratios, other accounting based measures and the risk- neutral distance metric from our structural model as explanatory variables we calibrate the hybrid model with an ordered – probit regression method. Using the same econometric method, we calibrate a model using financial ratios and accounting variables as explanatory variables and a model using our risk-neutral distance to default metric as unique explanatory variable. Then, using cumulative accuracy plots we have test the classification power of those models to predict defaults out of sample. We have found that by enriching the risk-neutral distance to default metric with financial ratios and accounting variables into the hybrid model, we can improve both in of-sample fit of credit ratings and out-of-sample predictability of defaults. Our main conclusion is that financial ratios and accounting variables contain significant and incremental information, thus the risk-neutral distance to default metric does not reflect all available information regarding the credit quality of a firm.
    Keywords: credit risk, distance to default, financial ratios, accounting variables
    JEL: G
    Date: 2005–05–27

This nep-dcm issue is ©2005 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.