Comparison of SML and GMM estimators for the random coefficient logit model using aggregate data

Sungho Park, Sachin Gupta

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

A simulated maximum likelihood (SML) estimator for the random coefficient logit model using aggregate data is found to be more efficient than the widely used generalized method of moments estimator (GMM) of Berry et al. (Econometrica 63:841-890, 1995). In particular, the SML estimator is better than the GMM estimator in recovery of heterogeneity parameters which are often of central interest in marketing research. With the GMM estimator, the analyst must determine what moment conditions to use for parameter identification, especially the heterogeneity parameters. With the SML estimator, the moment conditions are automatically determined as the gradients of the log-likelihood function, and these are the most efficient ones if the model is correctly specified. Another limitation of the GMM estimator is that the product market shares must be strictly positive while the SML estimator can handle zero market share observations. Properties of the SML and GMM estimators are demonstrated in simulated data and in data from the US photographic film market.

Original languageEnglish (US)
Pages (from-to)1353-1372
Number of pages20
JournalEmpirical Economics
Volume43
Issue number3
DOIs
StatePublished - Dec 2012

Fingerprint

Simulated Maximum Likelihood
Random Coefficient Models
Generalized Method of Moments
Moment Estimator
Logit Model
aggregate data
Maximum Likelihood Estimator
Estimator
Moment Conditions
market share
Strictly positive
Parameter Identification
Likelihood Function
marketing
Recovery
Simulated maximum likelihood
Generalized method of moments estimator
Logit model
Aggregate data
Random coefficients

Keywords

  • Endogeneity
  • Generalized method of moments
  • Heterogeneity
  • Logit model
  • Random coefficients
  • Simulated maximum likelihood

ASJC Scopus subject areas

  • Economics and Econometrics
  • Social Sciences (miscellaneous)
  • Mathematics (miscellaneous)
  • Statistics and Probability

Cite this

Comparison of SML and GMM estimators for the random coefficient logit model using aggregate data. / Park, Sungho; Gupta, Sachin.

In: Empirical Economics, Vol. 43, No. 3, 12.2012, p. 1353-1372.

Research output: Contribution to journalArticle

@article{0434791dfb674b2b9aad54ee2b2b480e,
title = "Comparison of SML and GMM estimators for the random coefficient logit model using aggregate data",
abstract = "A simulated maximum likelihood (SML) estimator for the random coefficient logit model using aggregate data is found to be more efficient than the widely used generalized method of moments estimator (GMM) of Berry et al. (Econometrica 63:841-890, 1995). In particular, the SML estimator is better than the GMM estimator in recovery of heterogeneity parameters which are often of central interest in marketing research. With the GMM estimator, the analyst must determine what moment conditions to use for parameter identification, especially the heterogeneity parameters. With the SML estimator, the moment conditions are automatically determined as the gradients of the log-likelihood function, and these are the most efficient ones if the model is correctly specified. Another limitation of the GMM estimator is that the product market shares must be strictly positive while the SML estimator can handle zero market share observations. Properties of the SML and GMM estimators are demonstrated in simulated data and in data from the US photographic film market.",
keywords = "Endogeneity, Generalized method of moments, Heterogeneity, Logit model, Random coefficients, Simulated maximum likelihood",
author = "Sungho Park and Sachin Gupta",
year = "2012",
month = "12",
doi = "10.1007/s00181-011-0519-3",
language = "English (US)",
volume = "43",
pages = "1353--1372",
journal = "Empirical Economics",
issn = "0377-7332",
publisher = "Physica-Verlag",
number = "3",

}

TY - JOUR

T1 - Comparison of SML and GMM estimators for the random coefficient logit model using aggregate data

AU - Park, Sungho

AU - Gupta, Sachin

PY - 2012/12

Y1 - 2012/12

N2 - A simulated maximum likelihood (SML) estimator for the random coefficient logit model using aggregate data is found to be more efficient than the widely used generalized method of moments estimator (GMM) of Berry et al. (Econometrica 63:841-890, 1995). In particular, the SML estimator is better than the GMM estimator in recovery of heterogeneity parameters which are often of central interest in marketing research. With the GMM estimator, the analyst must determine what moment conditions to use for parameter identification, especially the heterogeneity parameters. With the SML estimator, the moment conditions are automatically determined as the gradients of the log-likelihood function, and these are the most efficient ones if the model is correctly specified. Another limitation of the GMM estimator is that the product market shares must be strictly positive while the SML estimator can handle zero market share observations. Properties of the SML and GMM estimators are demonstrated in simulated data and in data from the US photographic film market.

AB - A simulated maximum likelihood (SML) estimator for the random coefficient logit model using aggregate data is found to be more efficient than the widely used generalized method of moments estimator (GMM) of Berry et al. (Econometrica 63:841-890, 1995). In particular, the SML estimator is better than the GMM estimator in recovery of heterogeneity parameters which are often of central interest in marketing research. With the GMM estimator, the analyst must determine what moment conditions to use for parameter identification, especially the heterogeneity parameters. With the SML estimator, the moment conditions are automatically determined as the gradients of the log-likelihood function, and these are the most efficient ones if the model is correctly specified. Another limitation of the GMM estimator is that the product market shares must be strictly positive while the SML estimator can handle zero market share observations. Properties of the SML and GMM estimators are demonstrated in simulated data and in data from the US photographic film market.

KW - Endogeneity

KW - Generalized method of moments

KW - Heterogeneity

KW - Logit model

KW - Random coefficients

KW - Simulated maximum likelihood

UR - http://www.scopus.com/inward/record.url?scp=84870334148&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870334148&partnerID=8YFLogxK

U2 - 10.1007/s00181-011-0519-3

DO - 10.1007/s00181-011-0519-3

M3 - Article

VL - 43

SP - 1353

EP - 1372

JO - Empirical Economics

JF - Empirical Economics

SN - 0377-7332

IS - 3

ER -