Semi-Parametric Analysis of Efficiency and Productivity using Gaussian Processes

Gregory Emvalomatis

doi:10.1093/ectj/utz013

Semi-Parametric Analysis of Efficiency and Productivity using Gaussian Processes

Gregory Emvalomatis (Lead / Corresponding author)

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

152 Downloads (Pure)

Abstract

This paper proposes a fully Bayesian semi-parametric method for efficiency and productivity analysis based on Gaussian processes. The proposed technique frees the researcher from having to specify a functional form for the production frontier, and it is shown in simulated data to perform as well as flexible parametric models when correct distributional assumptions are imposed on the inefficiency component of the error term, and slightly better when incorrect assumptions are made. The technique is applied to a panel dataset of US electric utilities, where total-factor productivity growth is estimated and decomposed with both parametric and semi-parametric techniques.

Original language	English
Pages (from-to)	48-67
Number of pages	20
Journal	Econometrics Journal
Volume	23
Issue number	1
Early online date	5 Sept 2019
DOIs	https://doi.org/10.1093/ectj/utz013
Publication status	Published - Jan 2020

Keywords

Gaussian-process regression
stochastic frontier
TFP decomposition
Gaussian process regression
total-factor productivity decomposition

ASJC Scopus subject areas

Aquatic Science

Access to Document

10.1093/ectj/utz013

Author Accepted Manuscript
This is a pre-copyedited, author-produced version of an article accepted for publication in [insert journal title] following peer review. The version of record [insert complete citation information here] is available online at: [insert URL and DOI of the article on the OUP website]
Accepted author manuscript, 568 KB

Cite this

@article{3631d879c39c4903bc5a345539ec60d7,

title = "Semi-Parametric Analysis of Efficiency and Productivity using Gaussian Processes",

abstract = "This paper proposes a fully Bayesian semi-parametric method for efficiency and productivity analysis based on Gaussian processes. The proposed technique frees the researcher from having to specify a functional form for the production frontier, and it is shown in simulated data to perform as well as flexible parametric models when correct distributional assumptions are imposed on the inefficiency component of the error term, and slightly better when incorrect assumptions are made. The technique is applied to a panel dataset of US electric utilities, where total-factor productivity growth is estimated and decomposed with both parametric and semi-parametric techniques.",

keywords = "Gaussian-process regression, stochastic frontier, TFP decomposition, Gaussian process regression, total-factor productivity decomposition",

author = "Gregory Emvalomatis",

year = "2020",

month = jan,

doi = "10.1093/ectj/utz013",

language = "English",

volume = "23",

pages = "48--67",

journal = "Econometrics Journal",

issn = "1368-4221",

publisher = "Oxford University Press",

number = "1",

}

TY - JOUR

T1 - Semi-Parametric Analysis of Efficiency and Productivity using Gaussian Processes

AU - Emvalomatis, Gregory

PY - 2020/1

Y1 - 2020/1

N2 - This paper proposes a fully Bayesian semi-parametric method for efficiency and productivity analysis based on Gaussian processes. The proposed technique frees the researcher from having to specify a functional form for the production frontier, and it is shown in simulated data to perform as well as flexible parametric models when correct distributional assumptions are imposed on the inefficiency component of the error term, and slightly better when incorrect assumptions are made. The technique is applied to a panel dataset of US electric utilities, where total-factor productivity growth is estimated and decomposed with both parametric and semi-parametric techniques.

AB - This paper proposes a fully Bayesian semi-parametric method for efficiency and productivity analysis based on Gaussian processes. The proposed technique frees the researcher from having to specify a functional form for the production frontier, and it is shown in simulated data to perform as well as flexible parametric models when correct distributional assumptions are imposed on the inefficiency component of the error term, and slightly better when incorrect assumptions are made. The technique is applied to a panel dataset of US electric utilities, where total-factor productivity growth is estimated and decomposed with both parametric and semi-parametric techniques.

KW - Gaussian-process regression

KW - stochastic frontier

KW - TFP decomposition

KW - Gaussian process regression

KW - total-factor productivity decomposition

UR - http://www.scopus.com/inward/record.url?scp=85084302588&partnerID=8YFLogxK

U2 - 10.1093/ectj/utz013

DO - 10.1093/ectj/utz013

M3 - Article

SN - 1368-4221

VL - 23

SP - 48

EP - 67

JO - Econometrics Journal

JF - Econometrics Journal

IS - 1

ER -

Semi-Parametric Analysis of Efficiency and Productivity using Gaussian Processes

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Erratum: Semi-parametric analysis of efficiency and productivity using Gaussian processes

Cite this