Bayesian and classical estimation of mixed logit: An application to genetic testing

Dean A. Regier, Mandy Ryan, Euan Phimister, Carlo A. Marra

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

Discrete choice experiments (DCEs) in health economics have recently Used the mixed logit (MXL) model to incorporate preference heterogeneity. These studies typically use a classical approach to estimation or have specified normal distributions for the attributes. Specifying normal distributions Call lead to erroneous interpretation: non-normal distributions may cause problems with convergence to the global maximum of the simulated log-likelihood function. Hierarchical Bayes (HB) of MXL is an alternative estimation approach that may alleviate problems of convergence. We investigated Bayesian and classical approaches to MXL estimation using a DCE that elicited preferences for a genetic technology. The classical approach produced Unrealistic results in one of the econometric specifications, which led to an erroneous willingness to pay estimate. The HB procedure produced reasonable results for both specifications and helped ascertain that the classical procedures were converging at a local maximum.

Original languageEnglish
Pages (from-to)598-610
Number of pages13
JournalJournal of Health Economics
Volume28
Issue number3
Early online date3 Dec 2008
DOIs
Publication statusPublished - May 2009

Keywords

  • genetic testing
  • discrete choice experiments
  • Hierarchical Bayes
  • maximum simulated likelihood estimation
  • discrete-choice experiments
  • chain monte-carlo
  • mental-retardation
  • developmental delay
  • models
  • heterogeneity
  • preferences
  • health
  • asthma

Cite this

Bayesian and classical estimation of mixed logit : An application to genetic testing. / Regier, Dean A.; Ryan, Mandy; Phimister, Euan; Marra, Carlo A.

In: Journal of Health Economics, Vol. 28, No. 3, 05.2009, p. 598-610.

Research output: Contribution to journalArticle

@article{53e5cf78605d49d8ac5db3d4255f44da,
title = "Bayesian and classical estimation of mixed logit: An application to genetic testing",
abstract = "Discrete choice experiments (DCEs) in health economics have recently Used the mixed logit (MXL) model to incorporate preference heterogeneity. These studies typically use a classical approach to estimation or have specified normal distributions for the attributes. Specifying normal distributions Call lead to erroneous interpretation: non-normal distributions may cause problems with convergence to the global maximum of the simulated log-likelihood function. Hierarchical Bayes (HB) of MXL is an alternative estimation approach that may alleviate problems of convergence. We investigated Bayesian and classical approaches to MXL estimation using a DCE that elicited preferences for a genetic technology. The classical approach produced Unrealistic results in one of the econometric specifications, which led to an erroneous willingness to pay estimate. The HB procedure produced reasonable results for both specifications and helped ascertain that the classical procedures were converging at a local maximum.",
keywords = "genetic testing, discrete choice experiments, Hierarchical Bayes, maximum simulated likelihood estimation, discrete-choice experiments, chain monte-carlo, mental-retardation, developmental delay, models, heterogeneity, preferences, health, asthma",
author = "Regier, {Dean A.} and Mandy Ryan and Euan Phimister and Marra, {Carlo A.}",
year = "2009",
month = "5",
doi = "10.1016/j.jhealeco.2008.11.003",
language = "English",
volume = "28",
pages = "598--610",
journal = "Journal of Health Economics",
issn = "0167-6296",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Bayesian and classical estimation of mixed logit

T2 - An application to genetic testing

AU - Regier, Dean A.

AU - Ryan, Mandy

AU - Phimister, Euan

AU - Marra, Carlo A.

PY - 2009/5

Y1 - 2009/5

N2 - Discrete choice experiments (DCEs) in health economics have recently Used the mixed logit (MXL) model to incorporate preference heterogeneity. These studies typically use a classical approach to estimation or have specified normal distributions for the attributes. Specifying normal distributions Call lead to erroneous interpretation: non-normal distributions may cause problems with convergence to the global maximum of the simulated log-likelihood function. Hierarchical Bayes (HB) of MXL is an alternative estimation approach that may alleviate problems of convergence. We investigated Bayesian and classical approaches to MXL estimation using a DCE that elicited preferences for a genetic technology. The classical approach produced Unrealistic results in one of the econometric specifications, which led to an erroneous willingness to pay estimate. The HB procedure produced reasonable results for both specifications and helped ascertain that the classical procedures were converging at a local maximum.

AB - Discrete choice experiments (DCEs) in health economics have recently Used the mixed logit (MXL) model to incorporate preference heterogeneity. These studies typically use a classical approach to estimation or have specified normal distributions for the attributes. Specifying normal distributions Call lead to erroneous interpretation: non-normal distributions may cause problems with convergence to the global maximum of the simulated log-likelihood function. Hierarchical Bayes (HB) of MXL is an alternative estimation approach that may alleviate problems of convergence. We investigated Bayesian and classical approaches to MXL estimation using a DCE that elicited preferences for a genetic technology. The classical approach produced Unrealistic results in one of the econometric specifications, which led to an erroneous willingness to pay estimate. The HB procedure produced reasonable results for both specifications and helped ascertain that the classical procedures were converging at a local maximum.

KW - genetic testing

KW - discrete choice experiments

KW - Hierarchical Bayes

KW - maximum simulated likelihood estimation

KW - discrete-choice experiments

KW - chain monte-carlo

KW - mental-retardation

KW - developmental delay

KW - models

KW - heterogeneity

KW - preferences

KW - health

KW - asthma

U2 - 10.1016/j.jhealeco.2008.11.003

DO - 10.1016/j.jhealeco.2008.11.003

M3 - Article

VL - 28

SP - 598

EP - 610

JO - Journal of Health Economics

JF - Journal of Health Economics

SN - 0167-6296

IS - 3

ER -