nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2013‒07‒15
nineteen papers chosen by
Edoardo Marcucci
Universita' di Roma Tre

  1. Stochastic Choice and Consideration Sets By Paola Manzini; Marco Mariotti
  2. Using Attribute Importance Rankings within Discrete Choice Experiments: An Application to Valuing Bread Attributes By Balcombe, Kelvin; Bitzios, Michael; Fraser, Iain; Haddock-Fraser, Janet
  3. A Behavioural Model of Choice in the Presence of Decision Conflict By Georgios Gerasimou
  4. Comparing models of unobserved heterogeneity in environmental choice experiments By Kragt, Marit E.
  5. Fitting Complex Mixed Logit Models with Particular Focus on Labor Supply Estimation By Max Löffler
  6. The Non-Market Value of Abel Tasman National Park, New Zealand: A Choice Modelling Application By Lee, Peter; Cassells, Sue; Holland, John
  7. Estimating the supply of on-farm biodiversity conservation services by north Australian pastoralists: design of a choice experiment By Greiner, Romy; Ballweg, Julie
  8. Farmer Preferences for Abiotic Stress Tolerance in Hybrid versus Inbred Rice: Evidence from Bihar, India By Ward, Patrick S.; Ortega, David L.; Spielman, David J.; Singh, Vartika; Magnan, Nicholas
  9. Using a Control Function to Resolve the Travel Cost Endogeneity Problem in Recreation Demand Models By Melstrom, Richard; Lupi, Frank
  10. The potential of food waste reduction through the green purchasing By Iwamoto, Hiroyuki
  11. Respiratory Health of Pacific Island Immigrants and Preferences for Indoor Air Quality Determinants in New Zealand By John Gibson; Riccardo Scarpa; Halahingano Rohorua
  12. Job Characteristics and Labour Supply By Lars Kunze; Nicolai Suppa
  13. Community values for the benefits of carbon farming: a choice experiment study By Massam, G.; Kragt, M.E.; Burton, M.
  14. The effects of experience on preference uncertainty: theory and empirics for environmental goods By Czajkowski, Mikołaj; Hanley, Nick; LaRiviere, Jacob
  15. Willingness to Pay for Private Labels, National Brands, and Local Designations at the Retail Level By Bosworth, Ryan C.; Bailey, DeeVon; Curtis, Kynda R.
  16. Multinomial and Mixed Logit Modeling in the Presence of Heterogeneity: A Two-Period Comparison of Healthcare Provider Choice in Rural China By Martine AUDIBERT; Yong HE; Jacky MATHONNAT
  17. Calibration of values in benefit transfer to account for variations in geographic scale and scope: Comparing two choice modelling experiments By Rolfe, John; Windle, Jill; Bennett, Jeff; Mazur, Kasia
  18. The impacts of the global food and financial crises on household food security and economic well-being: evidence from Bangladesh By Sonia, Akter; Syed Abul, Basher
  19. Endogenous variables in non-linear models with mixed effects: Inconsistence under perfect identification conditions? By Franz Buscha; Anna Conte

  1. By: Paola Manzini (University of St. Andrews); Marco Mariotti (University of St. Andrews)
    Abstract: We model a boundedly rational agent who suffers from limited attention. The agent considers each feasible alternative with a given (unobservable) probability,the attention parameter, and then chooses the alternative that maximises a preference relation within the set of considered alternatives. We show that this random choice rule is the only one for which the impact of removing an alternative on the choice probability of any other alternative is asymmetric and menu independent. Both the preference relation and the attention parameters are identified uniquely by stochastic choice data.
    Keywords: Discrete choice, Random utility, Logit model, Luce model, Consideration sets, bounded rationality, revealed preferences
    JEL: D0
    Date: 2013–07–09
  2. By: Balcombe, Kelvin; Bitzios, Michael; Fraser, Iain; Haddock-Fraser, Janet
    Abstract: We present a new Bayesian econometric speci…cation for a hypothetical Discrete Choice Experiment (DCE) incorporating respondent ranking information about attribute impor- tance. Our results indicate that a DCE debrie…ng question that asks respondents to rank the importance of attributes helps to explain the resulting choices. We also examine how mode of survey delivery (online and mail) impacts model performance, nding that results are not substantively a¤ected by the mode of survey delivery. We conclude that the ranking data is a complementary source of information about respondent utility functions within hypothetical DCEs.
    Keywords: Attribute Importance Rankings, Discrete Choice Experiment, Survey Mode, Food Consumption/Nutrition/Food Safety, Productivity Analysis, Research and Development/Tech Change/Emerging Technologies, C11, C25, L66,
    Date: 2013–02
  3. By: Georgios Gerasimou
    Abstract: This paper proposes a model of choice that does not assume completeness of the decision maker\'s preferences. The model explains in a natural way, and within a unied framework of choice when preference-incomparable options are present, four behavioural phenomena: the attraction effect, choice deferral, the strengthening of the attraction effect when deferral is permissible, and status quo bias. The key element in the proposed decision rule is that an individual chooses an alternative from a menu if it is worse than no other alternative in that menu and is also better than at least one. Utility-maximising behaviour is included as a special case when preferences are complete. The relevance of the partial dominance idea underlying the proposed choice procedure is illustrated with an intuitive generalisation of weakly dominated strategies and their iterated deletion in games with vector payoffs.
    Keywords: Choice with incomplete preferences; attraction effect; status quo bias; games with vector payoffs; iterative dominance
    JEL: D01
    Date: 2013–01–05
  4. By: Kragt, Marit E.
    Abstract: Choice experiments have become a widespread approach to non-market environmental valuation. Given the vast range of public opinions towards environmental management changes, it is desirable that analysis of discrete choice data accounts for the possibility of unobserved heterogeneity amongst the population. There is, however, no consensus about the best way to model individual heterogeneity. This paper presents four approaches to modelling heterogeneity that are increasingly used in the literature. Latent class, mixed logit, scaled multinational logit and generalised mixed logit (GMXL) models are estimated using case study data for catchment environmental management in Australia. A GMXL model that accounts for preference and scale heterogeneity performs best. I evaluate the impacts of models on welfare estimates and discuss the merits of each modelling approach.
    Keywords: Choice Modelling, Econometrics, Random Parameters, Scale Heterogeneity, Unobserved Preference Heterogeneity, Environmental Economics and Policy, Research and Development/Tech Change/Emerging Technologies, Research Methods/ Statistical Methods,
    Date: 2013–02
  5. By: Max Löffler (IZA, Bonn)
    Abstract: When one estimates discrete choice models, the mixed logit approach is commonly superior to simple conditional logit setups. Mixed logit models not only allow the researcher to implement difficult random components but also overcome the restrictive IIA assumption. Despite these theoretical advantages, the estimation of mixed logit models becomes cumbersome when the model’s complexity increases. Applied works therefore often rely on rather simple empirical specifications because this reduces the computational burden. I introduce the user-written command lslogit, which fits complex mixed logit models using maximum simulated likelihood methods. As lslogit is a d2-ML-evaluator written in Mata, the estimation is rather efficient compared with other routines. It allows the researcher to specify complicated structures of unobserved heterogeneity and to choose from a set of frequently used functional forms for the direct utility function—for example, Box-Cox transformations, which are difficult to estimate in the context of logit models. The particular focus of lslogit is on the estimation of labor supply models in the discrete choice context; therefore, it facilitates several computationally exhausting but standard tasks in this research area. However, the command can be used in many other applications of mixed logit models as well.
    Date: 2013–07–03
  6. By: Lee, Peter; Cassells, Sue; Holland, John
    Abstract: National parks and protected areas form the basis of global conservation initiatives and provide a raft of benefits in the form of various consumptive and non-consumptive uses. However, it is extremely difficult to express these benefits in monetary terms. The lack of economic values for these protected areas often results in sub-optimal conservation outcomes. Non-market valuation techniques can be used to estimate monetary values for these key environmental assets. This research applied the choice modelling approach to assess the value of non-market goods and services associated with Abel Tasman National Park in New Zealand. A standard multinomial logit model was used to analyse visitor preferences and derive welfare measures. The results indicate park users were willing to pay an actual cash value for the ecological and recreational attributes of the park. These monetary values can be used to guide future development, inform resource allocation decisions and ensure adequate conservation financing.
    Keywords: Choice experiments, stated preference, willing to pay, Consumer/Household Economics, Environmental Economics and Policy, International Relations/Trade, Research and Development/Tech Change/Emerging Technologies,
    Date: 2013–02
  7. By: Greiner, Romy; Ballweg, Julie
    Abstract: This paper reports on the experimental design process and considerations of a discretecontinuous choice experiment conducted in collaboration with landholders in northern Australia. The purpose of the research is to inform the design of effective and efficient payments-for-ecosystem services schemes to safeguard north Australia’s biodiversity values by promoting the contractual provision of biodiversity conservation services by landholders, in particular pastoralists and graziers. The paper focusses in particular on the discrete choice experimental (DCE) aspects. The DCE is employed to estimate landholders’ preference heterogeneity for supplying ecosystem services, specifically their willingness to accept remuneration for the on-farm conservation of biodiversity, based on potential program attributes. The design of the choice experiment draws on best practice standards (Hoyos 2010), a recognition of the benefits of embedding design in a consultative process (Klojgaard et al. 2012) and recent advances in accounting for response certainty (Brouwer et al. 2010; Hensher et al. 2012). DCE design decisions relating to attribute selection, attribute levels, alternatives and choice tasks are explained based on literature, focus group discussions, expert interviews and an iterative process of efficient DCE design. Additional design aspects include (i) a set of supplementary questions after each choice set to measure respondents’ choice certainty and elicit decision heuristics; (ii) embedding of the experiment in a socio-economic-psychological questionnaire, and (iii) logistical design.
    Keywords: Choice experimental design, efficient design, iterative process, response certainty, willingness to accept, farmers, on-farm biodiversity conservation, Agricultural and Food Policy, Environmental Economics and Policy, Farm Management, Institutional and Behavioral Economics,
    Date: 2013–02
  8. By: Ward, Patrick S.; Ortega, David L.; Spielman, David J.; Singh, Vartika; Magnan, Nicholas
    Abstract: Recent efforts to develop rice cultivars with drought-tolerance (DT) traits have resulted in the release of several varieties that demonstrate significant resiliency to drought stresses. In this paper, we use discrete choice experiments to examine farmers’ preferences for DT traits and explore heterogeneity in these preferences using primary data collected in rural Bihar, India. We evaluate farmers’ preference for yield performance under different weather scenarios, duration, seed reusability and seeding rate. Our results show that farmers value the reduction in yield variability offered by DT cultivars, but are willing to pay even more for cultivars that offer yield advantages even under normal conditions. Rice farmers were found to prefer short duration cultivars, which provide an alternative pathway by which farmers can manage drought risk. Finally, we find that farmers highly value seed-reusability, and would, other things equal, demand a discount on hybrid seeds that do not have this characteristic.
    Keywords: choice experiments, drought tolerance, rice, India, Agribusiness, Institutional and Behavioral Economics, International Development, International Relations/Trade, Research and Development/Tech Change/Emerging Technologies, Q12, Q16, O33,
    Date: 2013–06–03
  9. By: Melstrom, Richard; Lupi, Frank
    Abstract: This paper proposes using a control function to correct for endogeneity in recreation demand models. The control function approach is contrasted with the method of alternative specific constants (ASCs), which has been cautiously promoted in the literature. As an application, we consider the case of travel cost endogeneity in the demand for Great Lakes recreational fishing. Using data on Michigan anglers, we employ a random utility model of site choice. We show that either ASCs or the control function can correct for travel cost endogeneity, although we find that the model with ASCs produces significantly weaker results. Overall, compared with traditional approaches control functions may offer a more flexible means to eliminate endogeneity in recreation demand models.
    Keywords: Recreation demand, random utility model, travel cost method, travel cost endogeneity, control function, alternative specific constants, recreational fishing
    JEL: Q22 Q25 Q26
    Date: 2012–08
  10. By: Iwamoto, Hiroyuki
    Abstract: The purpose of this study is to examine the potential of food waste reduction through the green purchasing in Japanese consumers’ tofu purchasing decisions. Tofu is indispensable ingredient in Japanese cooking. But, huge amounts of food waste are produced during the paper production process. The Choice Modeling (Random Parameter Logit Model) is used in order to quantify the welfare change associated with the change in the level of local origin label, food recycling label, freshness of tofu, and price attribute for the sample of Japanese consumers in August 2012. The consumer has a positive perception of local origin label, food recycling label, freshness of tofu. The choice probability of food recycling labeled tofu is estimated at approximately 70%. The results suggest that green purchasing holds potential for food waste reduction in tofu manufacturing sector
    Keywords: choice experiment, food recycling, green purchase, Consumer/Household Economics, Environmental Economics and Policy, Food Consumption/Nutrition/Food Safety,
    Date: 2013–02
  11. By: John Gibson (University of Waikato); Riccardo Scarpa (University of Waikato); Halahingano Rohorua (University of Waikato)
    Abstract: Indoor air quality affects respiratory diseases, such as asthma, and can be altered by devices that lower dwelling humidity and raise temperature. Several countries have initiated schemes that subsidize devices such as heat pumps based on putative health benefits but the valuations of these devices by the affected populations remains unknown. We investigate preferences for devices that affect indoor air quality, dampness, and warmth, using a choice experiment with a sample of Pacific Islander immigrants in New Zealand. This is a high risk group for respiratory disease, who typically rent crowded and inadequately heated dwellings. Using both conditional logit and panel mixed logit models we find reasonably precise estimates of the willingness to pay for four improved heating and humidity control devices, which would cover the capital costs of two of the devices, and add up to about three-quarters of the cost of the other two devices.
    Keywords: respiratory health; indoor air-quality devices; choice experiments
    Date: 2013–06–30
  12. By: Lars Kunze; Nicolai Suppa
    Abstract: We document the importance of non-pecuniary aspects in employment relationships by showing that labour supply elasticities differ significantly among individuals’ job characteristics. Factor analysis indicates the relevance of four characteristics: autonomy, workload, variety and job security. Using a discrete choice model of family labour supply on the basis of Australian data, we show that income elasticities are significantly higher among individuals with “good” characteristics (e.g. a securer job) whereas wage elasticities are significantly lower. This result holds for both men and women. Our main hypothesis are derived within the ‘new approach to consumer theory proposed by Lancaster.
    Keywords: Labour supply; Discrete choice model; Job characteristics
    JEL: J22 J28 J32 C25
    Date: 2013–06
  13. By: Massam, G.; Kragt, M.E.; Burton, M.
    Abstract: The Australian Government’s Carbon Farming Initiative provides carbon credit incentives for farmers to encourage climate change mitigation on agricultural land. In addition to carbon sequestration or reduced emissions, carbon farming activities often generate ancillary benefits, such as creation of native habitat or erosion prevention. We conduct a choice experiment study to estimate community values for climate change mitigation, and the ancillary effects of carbon farming. Respondents’ WTP depends on their perceptions of climate change and on age, income and political preferences. Respondents who believe in climate change are willing to pay $7.56 per 1% reduction in Australia’s overall greenhouse gas emissions. Respondents are willing to pay $16.88 per 1% increase in the area of native vegetation on farmland, and $2.89 per 1% reduction in soil erosion. The value estimates will allow for more targeted development of carbon farming policies.
    Keywords: Agriculture, ClimateChange Mitigation, Carbon Farming, Choice Modelling, Environmental Economics and Policy, Farm Management, International Relations/Trade, Resource /Energy Economics and Policy,
    Date: 2013–02
  14. By: Czajkowski, Mikołaj; Hanley, Nick; LaRiviere, Jacob
    Abstract: This paper develops a model of demand estimation in which consumers learn about their true preferences through consumption experiences. We develop a theoretical model of Bayesian updating, perform comparative statics over the model, and show how the theoretical model can be consistently incorporated into a reduced form econometric model. We then estimate the model using data collected for two quasi-public goods. We find that the predictions of the theoretical exercise that additional experience with a good will make consumers more certain over their preferences in both mean and variance are supported in each case.
    Keywords: Bayesian, demand estimation, stated preference, generalized multinomial logit, scale, scale variance, Environmental Economics and Policy, Institutional and Behavioral Economics, Research and Development/Tech Change/Emerging Technologies, C51, D83, Q51, H43,
    Date: 2013–02
  15. By: Bosworth, Ryan C.; Bailey, DeeVon; Curtis, Kynda R.
    Abstract: A shopper survey was conducted to determine willingness to pay for ice cream with different labels. A statistical analysis was conducted using stated choices by respondents. The randomized choices were 1) a local brand with or without an indication it had a Utah’s Own designation, 2) a local brand with and without a locally-produced designation, 3) a private label product , and 4) a national brand product. The results suggested that brands affect willingness to pay for ice cream. However, shoppers were willing to pay a significant positive amount more for ice cream with the local designations.
    Keywords: choice experiments, state-sponsored food designations, ice cream, Consumer/Household Economics, Food Consumption/Nutrition/Food Safety, Public Economics, Q1, Q13,
    Date: 2013–02
  16. By: Martine AUDIBERT; Yong HE; Jacky MATHONNAT (Centre d'Etudes et de Recherches sur le Développement International)
    Abstract: This study aims at testing the theoretical issue according to which multinomial logit (MNL) would give lower performance than a mixed multinomial logit (MMNL) in the presence of heterogeneity. To do so, we construct two samples of patients surveyed within the same regions in rural China, but of an interval of 18 years, with a difference in preference heterogeneity due to income growth and population aging. With the 1989-1993 sample, both models have predicted price effects; however with the 2004-2006 sample, unlike MMNL, MNL failed to predict price effect. The explanation is that the impact of price on choice became more heterogeneous in the later than the former sample, thus heterogeneity makes a difference between MNL and MMNL. The absence of meaningful divergences of distance effects between the two models can also be explained by the evolution of heterogeneity in distance preferences over the period. The coefficients of price and distances with MMNL are higher than with MNL, indicating stronger price and distance effects in MMNL estimations. Another advantage of MMNL is the possibility to measure the extent of heterogeneity. The findings suggest caution when interpreting estimation results with MNL if heterogeneity is deemed important.
    Keywords: multinomial and mixed logit model, preference heterogeneity, healthcare choice, Chinese rural households
    JEL: D1 C5 I1
    Date: 2013
  17. By: Rolfe, John; Windle, Jill; Bennett, Jeff; Mazur, Kasia
    Abstract: Two choice modelling studies in Australia were designed to test for the effects of variations in geographic scale and scope on WTP values. One case study assessed values for improved natural resource management in a river catchment, and the other assessed values for improved protection of the Great Barrier Reef. The results show that increases in the amount of an amenity offered are valued positively and display diminishing marginal utility. Unit value estimates vary inversely with increases in the geographic scope over which an amenity improvement was offered. In the case studies, marginal values for the same unit of environmental improvement could be several thousand times higher when only very small areas were considered compared to when the whole amenity was framed. These results confirm that calibration factors are needed in benefit transfer applications between different geographic scopes. A close inverse relationship was identified between the ratio of quantities involved and the ratio of the WTP amounts. A log-log form of this relationship is recommended as a simple and efficient way of performing this calibration.
    Keywords: Demand and Price Analysis, Land Economics/Use, Research Methods/ Statistical Methods,
    Date: 2013–02
  18. By: Sonia, Akter; Syed Abul, Basher
    Abstract: This paper presents the first household-level study to examine the combined impacts of the global food and financial crises on household food security and economic well-being in a developing country. Using longitudinal survey data of 1,800 rural households from 12 districts of Bangladesh over the period 2007–2010, we estimated a three-stage hierarchical logit model to identify the key sources of household food insecurity. A difference-in-difference estimator was then employed to compare pre- and post-crises expenditure for those households who experienced acute food shortages and those who managed to avoid the worst impacts of the crises. On the basis of our results we conclude that: (1) the soaring food prices of 2007–2008 unequivocally aggravated food insecurity in the rural areas of Bangladesh in 2008; (2) there was some weak evidence to suggest that the global economic downturn, which followed the global food crisis, contributed towards worsening food insecurity in 2009; (3) the adverse impacts of these crises appeared to have faded over time due to labor and commodity market adjustments, regional economic growth, and domestic policy responses, leaving no profound, long-lasting impacts on households’ economic well-being; and (4) although the immediate adverse consequences of rising food prices were borne disproportionately by the poor and farming communities, the longer term consequences were distributed more evenly across the rich and poor and, in general, were favorable for the farming community.
    Keywords: Food security; Food price shocks; financial crises; Discrete choice modeling; Household survey data; Economic welfare; Bangladesh
    JEL: C25 I31 Q18
    Date: 2013–06–27
  19. By: Franz Buscha (Westminster Business School, University of Westminster); Anna Conte (Strategic Interaction Group, Max Planck Institute of Economics, Jena)
    Abstract: This paper examines the consequences of introducing a normally distributed effect into a system where the dependent variable is ordered and the explanatory variable is ordered and endogenous. Using simulation techniques we show that a naïve bivariate ordered probit estimator which fails to take a mixed effect into account will result in inconsistent estimates even when identification conditions are optimal. Our results suggest this finding only applies to non-linear endogenous systems.
    Keywords: bivariate probit, bivariate ordered probit, mixed effects, endogenous binary variables, constant parameters
    JEL: C35 C36 C51
    Date: 2013–07–01

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.