Masking identification of discrete choice models under simulation methods. (English) Zbl 1418.62430
Summary: We present examples based on actual and synthetic datasets to illustrate how simulation methods can mask identification problems in the estimation of discrete choice models such as mixed logit. Simulation methods approximate an integral (without a closed form) by taking draws from the underlying distribution of the random variable of integration. Our examples reveal how a low number of draws can generate estimates that appear identified, but in fact, are either not theoretically identified by the model or not empirically identified by the data. For the particular case of maximum simulated likelihood estimation, we investigate the underlying source of the problem by focusing on the shape of the simulated log-likelihood function under different conditions.
MSC:
| 62P20 | Applications of statistics to economics |
Software:
BIOGEMEReferences:
| [1] | Ben-Akiva, M., Bolduc, D., 1996. Multinomial probit with a logit kernel and a general parametric specification of the covariance structure. Working paper, Massachusetts Institute of Technology.; Ben-Akiva, M., Bolduc, D., 1996. Multinomial probit with a logit kernel and a general parametric specification of the covariance structure. Working paper, Massachusetts Institute of Technology. |
| [2] | Berndt, E.; Hall, B.; Hall, R.; Hausman, J., Estimation and inference in nonlinear structural models, Annals of Economics and Social Measurement, 3/4, 653-665 (1974) |
| [3] | Berry, S.; Levinsohn, J.; Pakes, A., Automobile prices in market equilibrium, Econometrica, 63, 4, 841-890 (1995) · Zbl 0836.90057 |
| [4] | Bhat, C., Quasi-random maximum simulated likelihood estimation of the mixed multinomial logit model, Transportation Research B, 35, 677-693 (2001) |
| [5] | Bierlaire, M., Bolduc, D., Godbout, M.-H., 2004. An introduction to BIOGEME (Version 1.0). Working paper, Ecole Polytechnique Federale de Lausanne.; Bierlaire, M., Bolduc, D., Godbout, M.-H., 2004. An introduction to BIOGEME (Version 1.0). Working paper, Ecole Polytechnique Federale de Lausanne. |
| [6] | Börsch-Supan, A.; Hajivassiliou, V., Smooth unbiased multivariate probability simulators for maximum likelihood estimation of limited dependent variable models, Journal of Econometrics, 58, 347-368 (1993) · Zbl 0798.62105 |
| [7] | Brownstone, D.; Train, K., Forecasting new product penetration with flexible substitution patterns, Journal of Econometrics, 89, 109-129 (1999) · Zbl 0962.62114 |
| [8] | Chiou, L., 2005. Empirical analysis of retail competition: spatial differentiation at Wal-Mart, Amazon.com, and their competitors. Working paper, Occidental College.; Chiou, L., 2005. Empirical analysis of retail competition: spatial differentiation at Wal-Mart, Amazon.com, and their competitors. Working paper, Occidental College. |
| [9] | Goolsbee, A.; Petrin, A., The consumer gains from direct broadcast satellites and the competition with cable television, Econometrica, 72, 351-381 (2003) · Zbl 1137.91518 |
| [10] | Halton, J., On the efficiency of evaluating certain quasi-random sequences of points in evaluating multi-dimensional integrals, Numerische Mathematik, 2, 84-90 (1960) · Zbl 0090.34505 |
| [11] | Hensher, D.; Greene, W., The mixed logit model: the state of practice, Transportation, 30, 133-176 (2003) |
| [12] | Hess, S., Polak, J.W., 2003a. A comparison of scrambled and shuffled Halton sequences for simulation based estimation. Centre for Transport Studies Working paper, Centre for Transport Studies, Imperial College London.; Hess, S., Polak, J.W., 2003a. A comparison of scrambled and shuffled Halton sequences for simulation based estimation. Centre for Transport Studies Working paper, Centre for Transport Studies, Imperial College London. |
| [13] | Hess, S., Polak, J.W., 2003b. An alternative to the scrambled Halton sequence for removing correlation between standard Halton sequences in high dimensions. Paper presented at the European Regional Science Conference, Jyväskylä.; Hess, S., Polak, J.W., 2003b. An alternative to the scrambled Halton sequence for removing correlation between standard Halton sequences in high dimensions. Paper presented at the European Regional Science Conference, Jyväskylä. |
| [14] | McFadden, D.; Train, K., Mixed MNL models of discrete response, Journal of Applied Econometrics, 15, 447-470 (2000) |
| [15] | Train, K., 1999. Halton sequences for mixed logit. Working paper, University of California, Berkeley.; Train, K., 1999. Halton sequences for mixed logit. Working paper, University of California, Berkeley. |
| [16] | Train, K., Discrete choice methods with simulation (2003), Cambridge University Press: Cambridge University Press New York · Zbl 1047.62098 |
| [17] | Train, K.; McFadden, D.; Ben-Akiva, M., The demand for local telephone service: a fully discrete model of residential calling patterns and service choices, RAND Journal of Economics, 18, 1, 109-123 (1987) |
| [18] | Walker, J.L., 2001. Extended discrete choice models: integrated framework, flexible error structures, and latent variables. Ph.D. dissertation, Massachusetts Institute of Technology, Department of Civil and Environmental Engineering.; Walker, J.L., 2001. Extended discrete choice models: integrated framework, flexible error structures, and latent variables. Ph.D. dissertation, Massachusetts Institute of Technology, Department of Civil and Environmental Engineering. |
| [19] | Walker, J.L., Ben-Akiva, M., Bolduc, D., 2006. Identification of parameters in normal error components logit mixture models. Journal of Applied Econometrics, forthcoming.; Walker, J.L., Ben-Akiva, M., Bolduc, D., 2006. Identification of parameters in normal error components logit mixture models. Journal of Applied Econometrics, forthcoming. |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
