Estimation of multinomial logit models with unobserved heterogeneity using maximum simulated likelihood
In this paper, we suggest a Stata routine for multinomial logit mod- els with unobserved heterogeneity using maximum simulated likelihood based on Halton sequences. The purpose of this paper is twofold. First, we describe the technical implementation of the estimation routine and discuss its properties. Further, we compare our estimation routine with the Stata program gllamm, which solves integration by using Gauss-Hermite quadrature or adaptive quadrature. For the analysis, we draw on multilevel data about schooling. Our empirical findings show that the estimation techniques lead to approximately the same estimation results. The advantage of simulation over Gauss-Hermite quadrature is a marked reduction in computational time for integrals with higher dimensions. Adaptive quadrature leads to more stable results relative to the other integration methods. However, simulation is more time efficient. We find that maximum simulated like- lihood leads to estimation results with reasonable accuracy in roughly half the time required when using adaptive quadrature. Copyright 2006 by StataCorp LP.
Volume (Year): 6 (2006)
Issue (Month): 2 (June)
|Contact details of provider:|| Web page: http://www.stata-journal.com/|
|Order Information:||Web: http://www.stata-journal.com/subscription.html|
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- Lorenzo Cappellari & Stephen P. Jenkins, 2003.
"Multivariate probit regression using simulated maximum likelihood,"
United Kingdom Stata Users' Group Meetings 2003
10, Stata Users Group.
- Lorenzo Cappellari & Stephen P. Jenkins, 2003. "Multivariate probit regression using simulated maximum likelihood," Stata Journal, StataCorp LP, vol. 3(3), pages 278-294, September.
- Sophia Rabe-Hesketh & Anders Skrondal & Andrew Pickles, 2002. "Reliable estimation of generalized linear mixed models using adaptive quadrature," Stata Journal, StataCorp LP, vol. 2(1), pages 1-21, February.
- Kenneth Train, 2001.
"Halton Sequences for Mixed Logit,"
- Kenneth Train ., 2000. "Halton Sequences for Mixed Logit," Economics Working Papers E00-278, University of California at Berkeley.
- Train, Kenneth, 2000. "Halton Sequences for Mixed Logit," Department of Economics, Working Paper Series qt6zs694tp, Department of Economics, Institute for Business and Economic Research, UC Berkeley.
- Sophia Rabe-Hesketh & Anders Skrondal, 2012. "Multilevel and Longitudinal Modeling Using Stata, 3rd Edition," Stata Press books, StataCorp LP, edition 3, number mimus2, September.
- William W. Gould & Jeffrey Pitblado & Brian Poi, 2010. "Maximum Likelihood Estimation with Stata," Stata Press books, StataCorp LP, edition 4, number ml4, September.
- Kenneth Train, 2003.
"Discrete Choice Methods with Simulation,"
Online economics textbooks,
SUNY-Oswego, Department of Economics, number emetr2.
- Cameron,A. Colin & Trivedi,Pravin K., 2005. "Microeconometrics," Cambridge Books, Cambridge University Press, number 9780521848053, December.
- 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.
- repec:tsj:spbook:mimus is not listed on IDEAS
- Peter Haan, 2005. "State Dependence and Female Labor Supply in Germany: The Extensive and the Intensive Margin," Discussion Papers of DIW Berlin 538, DIW Berlin, German Institute for Economic Research.
When requesting a correction, please mention this item's handle: RePEc:tsj:stataj:v:6:y:2006:i:2:p:229-245. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Christopher F. Baum)or (Lisa Gilmore)
If references are entirely missing, you can add them using this form.