In this article, I provide an illustrative, step-by-step implementation of the expectation-maximization algorithm for the nonparametric estimation of mixed logit models. In particular, the proposed routine allows users to fit straightforwardly latent-class logit models with an increasing number of mass points so as to approximate the unobserved structure of the mixing distribution.
BhatC. R.2000. Automatic modeling methods for univariate series. In Handbook of Transport Modelling, ed. HensherD. A., and ButtonK. J., 71–90. Oxford: Elsevier.
2.
BoylesR. A.1983. On the convergence of the EM algorithm. Journal of the Royal Statistical Society, Series B45: 47–50.
3.
DempsterA. P., LairdN. M., and RubinD. B.1977. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Series B39: 1–38.
4.
GreeneW. H., and HensherD. A.2003. A latent class model for discrete choice analysis: Contrasts with mixed logit. Transportation Research, Part B37: 681–698.
5.
HoleA. R.2007. Fitting mixed logit models by using maximum simulated likelihood. Stata Journal7: 388–401.
6.
HuberJ., and TrainK.2001. On the similarity of classical and Bayesian estimates of individual mean partworths. Marketing Letters12: 259–269.
7.
McFaddenD.1974. Conditional logit analysis of qualitative choice behavior. In Frontiers in Econometrics, ed. ZerembkaP., 105–142. New York: Academic Press.
8.
McFaddenD., and TrainK.2000. Mixed MNL models for discrete response. Journal of Applied Econometrics15: 447–470.
9.
TrainK. E.2003. Discrete Choice Methods with Simulation.Cambridge: Cambridge University Press.
10.
TrainK. E.2008. EM algorithms for nonparametric estimation of mixing distributions. Journal of Choice Modelling1: 40–69.
11.
WeeksD. E., and LangeK.1989. Trials, tribulations, and triumphs of the EM algorithm in pedigree analysis. Journal of Mathematics Applied in Medicine & Biology6: 209–232.
12.
WuC. F. J.1983. On the convergence properties of the EM algorithm. Annals of Statistics11: 95–103.
