Mixed-logit models are currently the state of the art in discrete-choice modelling, and their estimation in various forms (in particular, mixing revealed-preference and stated-preference data) is becoming increasingly popular. Although the theory behind these models is fairly simple, the practical problems associated with their estimation with empirical data are still relatively unknown and certainly not solved to everybody's satisfaction. In this paper we use a stated-preference dataset—previously used to derive willingness to pay for reduction in atmospheric pollution and subjective values of time—to estimate random parameter mixed logit models with different estimation methods. We use our results to discuss in some depth the problems associated with the derivation of willingness to pay with this class of models.
Get full access to this article
View all access options for this article.
References
1.
AlgersSBergstromMDahlbergMDillenJ, 1999, “Mixed logit estimation of the value of travel time”, working paper, Department of Economics, Uppsala University, Uppsala.
2.
AllenbyGRossiP, 1999, “Marketing models for consumer heterogeneity”Journal of Econometrics8957–78.
3.
AndrewsR LAnsariACurrimI S, 2002, “Hierarchical Bayes versus finite mixture conjoint analysis models: A Comparison of fit, prediction and partworth recovery”Journal of Marketing Research3987–98.
4.
ArmstrongP MGarridoR AOrtúzarJ de D, 2001, “Confidence intervals to bound the value of time”Transportation Research37E143–161.
5.
AroraNAllenbyGGinterJ, 1998, “A hierarchical Bayes model of primary and secondary demand”Marketing Science1729–44.
6.
BhatC, 2001, “Quasi-random maximum simulated likelihood estimation of the mixed multinomial logit model”Transportation Research35B677–695.
7.
ChibSGreenbergE, 1995, “Understanding the Metropolis–Hastings algorithm”The American Statistician49327–335.
8.
CowlesM KCarlinB P, 1996, “Markov chain Monte Carlo convergence diagnostics: A Comparative review”Journal of the American Statistical Association91883–904.
9.
EspinoROrtúzarJ de DRománC, 2004, “Confidence intervals for willingness to pay measures in mode choice models”, in Proceedings XIII Panamerican Congress of Traffic and Transportation Engineering Albany, USA, http://www.eng.rpi.edu/panam/.
10.
EttemaDGunnHDe JongGLindveldK, 1997, “A simulation method for determining the confidence interval of a weighted group average value of time”, in Transportation Planning Methods I, Volume P414. Proceedings of the 25th European Transport Forum (PTRC Education and Research Services, London) pp 101–112.
11.
FiellerE, 1932, “The distribution of the index in a normal bivariate population”Biometrika24428–440.
12.
GalileaPOrtúzarJ de D, 2004, “Valuing noise level reductions in a residential location context”Transportation Research9D forthcoming.
13.
GarridoR ASilvaM, 2004, “Low discrepancy sequences for the estimation of mixed logit models”Transportation Research B forthcoming.
14.
GaudryMJIJara-DíazS ROrtúzarJ de D, 1989, “Value of time sensitivity to model specification”Transportation Research23B151–158.
15.
GodoyG, 2004, “Estimatión Bayesiana de modelos flexibles de electión discreta” [Bayesian estimation of flexible discrete choice models], MSc thesis, Department of Transport Engineering, Pontificia Universidad Católica de Chile, Santiago.
16.
HajivassiliouVRuudP, 1994, “Classical estimation methods for LDV models using simulation”, in Handbook of Econometrics, Volume IV Eds EngleRMcFaddenD (Elsevier, New York) pp 2383–2441.
17.
HensherD A, 2001a, “The sensitivity of the valuation of travel time savings to the specification of unobserved effects”Transportation Research37E129–142.
18.
HensherD A, 2001b, “The valuation of commuter travel time savings for car drivers: Evaluating alternative model specifications”Transportation28101–118.
19.
HensherD A, 2001c, “Measurement of the valuation of travel time savings”Journal of Transport Economics and Policy3571–98.
20.
HensherD AGreeneW H, 2003, “The mixed logit model: The state of practice”Transportation30133–176.
21.
HinkleyD, 1969, “On the ratio of two correlated normal random variables”Biometrika56635–639.
22.
HorowitzJ L, 1981, “Sampling error, specification and data errors in probabilistic discrete choice models”, appendix C of Applied Discrete Choice Modelling D A Hensher, L W Johnson (Croom Helm, London) pp 417–435.
23.
HuberJTrainK E, 2001, “On the similarity of classical and Bayesian estimates of individual mean partworths”Marketing Letters12257–267.
24.
IragüenPOrtúzarJ de D, 2004, “Willingness-to-pay for reducing fatal accident risk in urban areas: An Internet-based web page stated preference survey”Accident Analysis and Prevention36513–524.
25.
Jara-DíazS R, 1990, “Income and taste in mode choice models: Are they surrogates?”Transportation Research25B341–350.
26.
KassRCarlinBGelmanANealR, 1998, “Markov chain Monte Carlo in practice: A Roundtable discussion”The American Statistician5293–100.
27.
LahiriKGaoJ, 2001, “Bayesian analysis of nested logit model by Markov chain Monte Carlo”, working paper, Department of Economics, State University of New York at Albany, NY.
28.
LenkPDe SarboWGreenPYoungM, 1999, “Hierarchical Bayes conjoint analysis: Recovery of partworth heterogeneity from reduced experimental designs”Marketing Science15173–191.
29.
LouviereJ JHensherD ASwaitJ D, 2000Stated Choice Methods: Analysis and Applications (Cambridge University Press, Cambridge).
30.
McCullochRRossiP, 1994, “An exact likelihood analysis of the multinomial probit model”Journal of Econometrics64207–240.
31.
McFaddenDTrainK E, 2000, “Mixed MNL models for discrete response”Journal of Applied Econometrics15447–470.
32.
MeijerERouwendalJ, 2000, “Measuring welfare effects in models with random coefficients”, research report 00F25, SOM Research School, University of Groningen, Groningen.
33.
MoreyERossmanK G, 2002, “Using stated-preference questions to investigate variation in willingness to pay for preserving marble monuments: classical heterogeneity and random parameters”, WP, Economics Department, University of Colorado at Boulder, CO.
34.
OrtúzarJ de DRodríguezG, 2002, “Valuing reductions in environmental pollution in a residential location context”Transportation Research7D407–427.
OrtúzarJ de DMartínezF JVarelaF J, 2000, “Stated preferences in modelling accessibility”International Planning Studies565–85.
37.
OrtúzarJ de DRodríguezGSillanoM, 2002, “Willingness-to-pay for reducing atmospheric pollution”Proceedings of the European Transport Conference, Homerton College, Cambridge, CD, PTRC Education and Research Services Ltd, London, http://www.ptrc-training.co.uk.
38.
PérezP EMartínezF JOrtúzarJ de D, 2003, “Microeconomic formulation and estimation of a residential location model: Implications for the value of time”Journal of Regional Science43771–789.
39.
RafteryALewisS, 1992, “How many iterations in the Gibbs sampler?”, in Bayesian Statistics 4 Eds BernardoJ MSmithAFMDavidA PBergerJ O (Oxford University Press, New York) pp 763–773.
40.
ReveltDTrainK E, 1998, “Mixed logit with repeated choices: Households’ choice of appliance efficiency level”Review of Economics and Statistics80647–657.
41.
ReveltDTrainK E, 2000, “Customer-specific taste parameters and mixed logit”, WP E00-274, Department of Economics, University of California at Berkeley, CA.
42.
RizziL IOrtúzarJ de D, 2003, “Stated preference in the valuation of interurban road safety”Accident Analysis and Prevention359–22.
43.
RizziL IOrtúzarJ de D, 2004, “Road safety valuation under a stated choice framework”Journal of Transport Economics and Policy in press.
44.
RuudP, 1996, “Simulation of the multinomial probit model: an analysis of covariance matrix estimation”, working paper, Department of Economics, University of California at Berkeley, CA.
SmallK ERosenH, 1981, “Applied welfare economics with discrete choice models”Econometrica49105–130.
47.
SpiegelhalterD JThomasABestN G, 2001Win BUGS Beta Version 1.4 User Manual MRC Biostatistics Unit, Institute of Public Health, University of Cambridge, Cambridge.
48.
TrainK E, 1998, “Recreational demand models with taste differences over people”Land Economics74230–239.
49.
TrainK E, 2001, “A comparison of hierarchical Bayes and maximum simulated likelihood for mixed logit”, working paper, Department of Economics, University of California at Berkeley, CA.
50.
TrainK E, 2003Discrete Choice Methods with Simulation (Cambridge University Press, Cambridge).
51.
TrainK ESonnierG, 2003, “Mixed logit with bounded distributions of partworths”, working paper, Department of Economics, University of California at Berkeley, CA.
52.
WalkerJ, 2001Extended Discrete Choice Models: Integrated Framework, Flexible Error Structures and Latent Variables PhD dissertation, Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, MA.
