IDEAS home Printed from https://ideas.repec.org/a/eee/eejocm/v31y2019icp73-85.html
   My bibliography  Save this article

Simulation error in maximum likelihood estimation of discrete choice models

Author

Listed:
  • Czajkowski, Mikołaj
  • Budziński, Wiktor

Abstract

Maximum simulated likelihood is the preferred estimator of most researchers who deal with discrete choice. It allows estimation of models such as mixed multinomial logit (MXL), generalized multinomial logit, or hybrid choice models, which have now become the state-of-practice in the microeconometric analysis of discrete choice data. All these models require simulation-based solving of multidimensional integrals, which can lead to several numerical problems. In this study, we focus on one of these problems – utilizing from 100 to 1,000,000 draws, we investigate the extent of the simulation bias resulting from using several different types of draws: (1) pseudo random numbers, (2) modified Latin hypercube sampling, (3) randomized scrambled Halton sequence, and (4) randomized scrambled Sobol sequence. Each estimation is repeated up to 1 000 times. The simulations use several artificial datasets based on an MXL data generating process with different numbers of individuals (400, 800, 1 200), different numbers of choice tasks per respondent (4, 8, 12), different number of attributes (5, 10), and different experimental designs (D-optimal, D-efficient for the MNL and D-efficient for the MXL model). Our large-scale simulation study allows for comparisons and drawing conclusions with respect to (1) how efficient different types of quasi Monte Carlo simulation methods are and (2) how many draws one should use to make sure the results are of “satisfying” quality – under different experimental conditions. Our study is the first to date to offer such a comprehensive comparison. Overall, we find that the number of the best-performing Sobol draws required for the desired precision exceeds 2 000 in some of the 5-attribute settings, and 20,000 in the case of some 10-attribute settings considered.

Suggested Citation

  • Czajkowski, Mikołaj & Budziński, Wiktor, 2019. "Simulation error in maximum likelihood estimation of discrete choice models," Journal of choice modelling, Elsevier, vol. 31(C), pages 73-85.
  • Handle: RePEc:eee:eejocm:v:31:y:2019:i:c:p:73-85
    DOI: 10.1016/j.jocm.2019.04.003
    as

    Download full text from publisher

    File URL: http://www.sciencedirect.com/science/article/pii/S1755534517302002
    Download Restriction: Full text for ScienceDirect subscribers only

    File URL: https://libkey.io/10.1016/j.jocm.2019.04.003?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to look for a different version below or search for a different version of it.

    Other versions of this item:

    References listed on IDEAS

    as
    1. Sándor, Zsolt & Train, Kenneth, 2004. "Quasi-random simulation of discrete choice models," Transportation Research Part B: Methodological, Elsevier, vol. 38(4), pages 313-327, May.
    2. Czajkowski, Mikolaj & Scasný, Milan, 2010. "Study on benefit transfer in an international setting. How to improve welfare estimates in the case of the countries' income heterogeneity?," Ecological Economics, Elsevier, vol. 69(12), pages 2409-2416, October.
    3. Chiou, Lesley & Walker, Joan L., 2007. "Masking identification of discrete choice models under simulation methods," Journal of Econometrics, Elsevier, vol. 141(2), pages 683-703, December.
    4. Fosgerau, Mogens & Mabit, Stefan L., 2013. "Easy and flexible mixture distributions," Economics Letters, Elsevier, vol. 120(2), pages 206-210.
    5. Laura Mørch Andersen, 2014. "Obtaining Reliable Likelihood Ratio Tests from Simulated Likelihood Functions," PLOS ONE, Public Library of Science, vol. 9(10), pages 1-12, October.
    6. Kenneth Train ., 2000. "Halton Sequences for Mixed Logit," Economics Working Papers E00-278, University of California at Berkeley.
    7. Daniel McFadden & Kenneth Train, 2000. "Mixed MNL models for discrete response," Journal of Applied Econometrics, John Wiley & Sons, Ltd., vol. 15(5), pages 447-470.
    8. Czajkowski, Mikołaj & Ahtiainen, Heini & Artell, Janne & Meyerhoff, Jürgen, 2017. "Choosing a Functional Form for an International Benefit Transfer: Evidence from a Nine-country Valuation Experiment," Ecological Economics, Elsevier, vol. 134(C), pages 104-113.
    9. Fosgerau, Mogens & Bierlaire, Michel, 2007. "A practical test for the choice of mixing distribution in discrete choice models," Transportation Research Part B: Methodological, Elsevier, vol. 41(7), pages 784-794, August.
    10. Bhat, Chandra R., 2001. "Quasi-random maximum simulated likelihood estimation of the mixed multinomial logit model," Transportation Research Part B: Methodological, Elsevier, vol. 35(7), pages 677-693, August.
    11. Train,Kenneth E., 2009. "Discrete Choice Methods with Simulation," Cambridge Books, Cambridge University Press, number 9780521766555.
    12. David Revelt & Kenneth Train, 1998. "Mixed Logit With Repeated Choices: Households' Choices Of Appliance Efficiency Level," The Review of Economics and Statistics, MIT Press, vol. 80(4), pages 647-657, November.
    13. Russell Davidson & James MacKinnon, 2000. "Bootstrap tests: how many bootstraps?," Econometric Reviews, Taylor & Francis Journals, vol. 19(1), pages 55-68.
    14. Laura Mørch Andersen, 2013. "Obtaining reliable Likelihood Ratio tests from simulated likelihood functions," IFRO Working Paper 2013/1, University of Copenhagen, Department of Food and Resource Economics.
    15. Hess, Stephane & Train, Kenneth E. & Polak, John W., 2006. "On the use of a Modified Latin Hypercube Sampling (MLHS) method in the estimation of a Mixed Logit Model for vehicle choice," Transportation Research Part B: Methodological, Elsevier, vol. 40(2), pages 147-163, February.
    16. Bhat, Chandra R., 2003. "Simulation estimation of mixed discrete choice models using randomized and scrambled Halton sequences," Transportation Research Part B: Methodological, Elsevier, vol. 37(9), pages 837-855, November.
    17. Riccardo Scarpa & John M. Rose, 2008. "Design efficiency for non-market valuation with choice modelling: how to measure it, what to report and why ," Australian Journal of Agricultural and Resource Economics, Australian Agricultural and Resource Economics Society, vol. 52(3), pages 253-282, September.
    18. Gregory L. Poe & Kelly L. Giraud & John B. Loomis, 2005. "Computational Methods for Measuring the Difference of Empirical Distributions," American Journal of Agricultural Economics, Agricultural and Applied Economics Association, vol. 87(2), pages 353-365.
    19. Matthew C. Harding & Jerry Hausman, 2007. "Using A Laplace Approximation To Estimate The Random Coefficients Logit Model By Nonlinear Least Squares," International Economic Review, Department of Economics, University of Pennsylvania and Osaka University Institute of Social and Economic Research Association, vol. 48(4), pages 1311-1328, November.
    20. Nick Hanley & Mikolaj Czajkowski, 2017. "Stated Preference Valuation Methods: An Evolving Tool for Understanding Choices and Informing Policy," Discussion Papers in Environment and Development Economics 2017-01, University of St. Andrews, School of Geography and Sustainable Development.
    21. Bliemer, Michiel C.J. & Rose, John M., 2010. "Construction of experimental designs for mixed logit models allowing for correlation across choice observations," Transportation Research Part B: Methodological, Elsevier, vol. 44(6), pages 720-734, July.
    22. Bhat, Chandra R. & Sidharthan, Raghuprasad, 2011. "A simulation evaluation of the maximum approximate composite marginal likelihood (MACML) estimator for mixed multinomial probit models," Transportation Research Part B: Methodological, Elsevier, vol. 45(7), pages 940-953, August.
    23. Heidi Tuhkanen & Evelin Urbel-Piirsalu & Tea Nõmmann & Mikołaj Czajkowski & Nick Hanley, 2015. "Valuing the benefits of improved marine environmental quality under multiple stressors," Working Papers 2015-41, Faculty of Economic Sciences, University of Warsaw.
    24. Dadi Kristofersson & Ståle Navrud, 2005. "Validity Tests of Benefit Transfer – Are We Performing the Wrong Tests?," Environmental & Resource Economics, Springer;European Association of Environmental and Resource Economists, vol. 30(3), pages 279-286, March.
    25. Fabian Bastin & Cinzia Cirillo & Philippe L. Toint, 2010. "Estimating Nonparametric Random Utility Models with an Application to the Value of Time in Heterogeneous Populations," Transportation Science, INFORMS, vol. 44(4), pages 537-549, November.
    26. Munger, D. & L’Ecuyer, P. & Bastin, F. & Cirillo, C. & Tuffin, B., 2012. "Estimation of the mixed logit likelihood function by randomized quasi-Monte Carlo," Transportation Research Part B: Methodological, Elsevier, vol. 46(2), pages 305-320.
    27. Train, Kenneth, 2016. "Mixed logit with a flexible mixing distribution," Journal of choice modelling, Elsevier, vol. 19(C), pages 40-53.
    28. Patrick Bajari & Jeremy T. Fox & Stephen P. Ryan, 2007. "Linear Regression Estimation of Discrete Choice Models with Nonparametric Distributions of Random Coefficients," American Economic Review, American Economic Association, vol. 97(2), pages 459-463, May.
    29. Richard T. Carson & Miko_aj Czajkowski, 2014. "The discrete choice experiment approach to environmental contingent valuation," Chapters, in: Stephane Hess & Andrew Daly (ed.), Handbook of Choice Modelling, chapter 9, pages 202-235, Edward Elgar Publishing.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Cherchi, Elisabetta & Guevara, Cristian Angelo, 2012. "A Monte Carlo experiment to analyze the curse of dimensionality in estimating random coefficients models with a full variance–covariance matrix," Transportation Research Part B: Methodological, Elsevier, vol. 46(2), pages 321-332.
    2. Staus, Alexander, 2008. "Standard and Shuffled Halton Sequences in a Mixed Logit Model," Working Papers 93856, Universitaet Hohenheim, Institute of Agricultural Policy and Agricultural Markets.
    3. Tinessa, Fiore & Marzano, Vittorio & Papola, Andrea, 2020. "Mixing distributions of tastes with a Combination of Nested Logit (CoNL) kernel: Formulation and performance analysis," Transportation Research Part B: Methodological, Elsevier, vol. 141(C), pages 1-23.
    4. Stephane Hess, 2014. "Latent class structures: taste heterogeneity and beyond," Chapters, in: Stephane Hess & Andrew Daly (ed.), Handbook of Choice Modelling, chapter 14, pages 311-330, Edward Elgar Publishing.
    5. Paleti, Rajesh, 2018. "Generalized multinomial probit Model: Accommodating constrained random parameters," Transportation Research Part B: Methodological, Elsevier, vol. 118(C), pages 248-262.
    6. Bansal, Prateek & Daziano, Ricardo A. & Achtnicht, Martin, 2018. "Comparison of parametric and semiparametric representations of unobserved preference heterogeneity in logit models," Journal of choice modelling, Elsevier, vol. 27(C), pages 97-113.
    7. Rico Krueger & Akshay Vij & Taha H. Rashidi, 2018. "A Dirichlet Process Mixture Model of Discrete Choice," Papers 1801.06296, arXiv.org.
    8. Marco A. Palma & Dmitry V. Vedenov & David Bessler, 2020. "The order of variables, simulation noise, and accuracy of mixed logit estimates," Empirical Economics, Springer, vol. 58(5), pages 2049-2083, May.
    9. Riccardo Scarpa & Cristiano Franceschinis & Mara Thiene, 2017. "A Monte Carlo Evaluation of the Logit-Mixed Logit under Asymmetry and Multimodality," Working Papers in Economics 17/23, University of Waikato.
    10. Akshay Vij & Rico Krueger, 2018. "Random taste heterogeneity in discrete choice models: Flexible nonparametric finite mixture distributions," Papers 1802.02299, arXiv.org.
    11. Vij, Akshay & Krueger, Rico, 2017. "Random taste heterogeneity in discrete choice models: Flexible nonparametric finite mixture distributions," Transportation Research Part B: Methodological, Elsevier, vol. 106(C), pages 76-101.
    12. Munger, D. & L’Ecuyer, P. & Bastin, F. & Cirillo, C. & Tuffin, B., 2012. "Estimation of the mixed logit likelihood function by randomized quasi-Monte Carlo," Transportation Research Part B: Methodological, Elsevier, vol. 46(2), pages 305-320.
    13. Bansal, Prateek & Daziano, Ricardo A. & Achtnicht, Martin, 2018. "Extending the logit-mixed logit model for a combination of random and fixed parameters," Journal of choice modelling, Elsevier, vol. 27(C), pages 88-96.
    14. Czajkowski, Mikołaj & Bartczak, Anna & Giergiczny, Marek & Navrud, Stale & Żylicz, Tomasz, 2014. "Providing preference-based support for forest ecosystem service management," Forest Policy and Economics, Elsevier, vol. 39(C), pages 1-12.
    15. West, Grant H. & Snell, Heather & Kovacs, Kent & Nayga, Rodolfo M., 2020. "Estimation of the preferences for the intertemporal services from groundwater," 2020 Annual Meeting, July 26-28, Kansas City, Missouri 304220, Agricultural and Applied Economics Association.
    16. Bliemer, Michiel C.J. & Rose, John M., 2010. "Construction of experimental designs for mixed logit models allowing for correlation across choice observations," Transportation Research Part B: Methodological, Elsevier, vol. 44(6), pages 720-734, July.
    17. Junyi Shen, 2009. "Latent class model or mixed logit model? A comparison by transport mode choice data," Applied Economics, Taylor & Francis Journals, vol. 41(22), pages 2915-2924.
    18. Hess, Stephane & Train, Kenneth E., 2011. "Recovery of inter- and intra-personal heterogeneity using mixed logit models," Transportation Research Part B: Methodological, Elsevier, vol. 45(7), pages 973-990, August.
    19. Yu, Jie & Goos, Peter & Vandebroek, Martina, 2010. "Comparing different sampling schemes for approximating the integrals involved in the efficient design of stated choice experiments," Transportation Research Part B: Methodological, Elsevier, vol. 44(10), pages 1268-1289, December.
    20. Prateek Bansal & Vahid Keshavarzzadeh & Angelo Guevara & Shanjun Li & Ricardo A Daziano, 2022. "Designed quadrature to approximate integrals in maximum simulated likelihood estimation [Evaluating simulation-based approaches and multivariate quadrature on sparse grids in estimating multivariat," The Econometrics Journal, Royal Economic Society, vol. 25(2), pages 301-321.

    More about this item

    Keywords

    Discrete choice; Mixed logit; Simulated maximum log-likelihood function; Simulation error; Draws; Quasi Monte Carlo methods; MLHS; Halton; Sobol; Number of draws;
    All these keywords.

    JEL classification:

    • C15 - Mathematical and Quantitative Methods - - Econometric and Statistical Methods and Methodology: General - - - Statistical Simulation Methods: General
    • C51 - Mathematical and Quantitative Methods - - Econometric Modeling - - - Model Construction and Estimation
    • C63 - Mathematical and Quantitative Methods - - Mathematical Methods; Programming Models; Mathematical and Simulation Modeling - - - Computational Techniques

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:eee:eejocm:v:31:y:2019:i:c:p:73-85. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Catherine Liu (email available below). General contact details of provider: http://www.journals.elsevier.com/journal-of-choice-modelling .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.