Bayesian and non-Bayesian analysis of gamma stochastic frontier models by Markov chain Monte Carlo methods

Hideo Kozumi; Xingyuan Zhang

doi:10.1007/BF02741316

Bayesian and non-Bayesian analysis of gamma stochastic frontier models by Markov chain Monte Carlo methods

Hideo Kozumi, Xingyuan Zhang

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

This paper considers simulation-based approaches for the gamma stochastic frontier model. Efficient Markov chain Monte Carlo methods are proposed for sampling the posterior distribution of the parameters. Maximum likelihood estimation is also discussed based on the stochastic approximation algorithm. The methods are applied to a data set of the U.S. electric utility industry.

Original language	English
Pages (from-to)	575-593
Number of pages	19
Journal	Computational Statistics
Volume	20
Issue number	4
DOIs	https://doi.org/10.1007/BF02741316
Publication status	Published - Dec 1 2006

Keywords

Acceptance-rejection Metropolis-Hastings algorithm
Auxiliary variable method
Marginal likelihood
Markov chain Monte Carlo
Stochastic approximation
Stochastic frontier model

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty
Computational Mathematics

Access to Document

10.1007/BF02741316

Cite this

@article{3874342bc96e43a2b0fae482f3f1c230,

title = "Bayesian and non-Bayesian analysis of gamma stochastic frontier models by Markov chain Monte Carlo methods",

abstract = "This paper considers simulation-based approaches for the gamma stochastic frontier model. Efficient Markov chain Monte Carlo methods are proposed for sampling the posterior distribution of the parameters. Maximum likelihood estimation is also discussed based on the stochastic approximation algorithm. The methods are applied to a data set of the U.S. electric utility industry.",

keywords = "Acceptance-rejection Metropolis-Hastings algorithm, Auxiliary variable method, Marginal likelihood, Markov chain Monte Carlo, Stochastic approximation, Stochastic frontier model",

author = "Hideo Kozumi and Xingyuan Zhang",

year = "2006",

month = dec,

day = "1",

doi = "10.1007/BF02741316",

language = "English",

volume = "20",

pages = "575--593",

journal = "Computational Statistics",

issn = "0943-4062",

publisher = "Springer Verlag",

number = "4",

}

TY - JOUR

T1 - Bayesian and non-Bayesian analysis of gamma stochastic frontier models by Markov chain Monte Carlo methods

AU - Kozumi, Hideo

AU - Zhang, Xingyuan

PY - 2006/12/1

Y1 - 2006/12/1

N2 - This paper considers simulation-based approaches for the gamma stochastic frontier model. Efficient Markov chain Monte Carlo methods are proposed for sampling the posterior distribution of the parameters. Maximum likelihood estimation is also discussed based on the stochastic approximation algorithm. The methods are applied to a data set of the U.S. electric utility industry.

AB - This paper considers simulation-based approaches for the gamma stochastic frontier model. Efficient Markov chain Monte Carlo methods are proposed for sampling the posterior distribution of the parameters. Maximum likelihood estimation is also discussed based on the stochastic approximation algorithm. The methods are applied to a data set of the U.S. electric utility industry.

KW - Acceptance-rejection Metropolis-Hastings algorithm

KW - Auxiliary variable method

KW - Marginal likelihood

KW - Markov chain Monte Carlo

KW - Stochastic approximation

KW - Stochastic frontier model

UR - http://www.scopus.com/inward/record.url?scp=37449025952&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=37449025952&partnerID=8YFLogxK

U2 - 10.1007/BF02741316

DO - 10.1007/BF02741316

M3 - Article

AN - SCOPUS:37449025952

SN - 0943-4062

VL - 20

SP - 575

EP - 593

JO - Computational Statistics

JF - Computational Statistics

IS - 4

ER -

Bayesian and non-Bayesian analysis of gamma stochastic frontier models by Markov chain Monte Carlo methods

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this