Home > Research > Publications & Outputs > A coherent approach to Bayesian Data Envelopmen...

Electronic data

  • paper_revised

    Rights statement: This is the author’s version of a work that was accepted for publication in European Journal of Operational Research. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in European Journal of Operational Research, 281, 2, 2020 DOI: 10.1016/j.ejor.2019.08.039

    Accepted author manuscript, 686 KB, PDF document

    Available under license: CC BY-NC-ND

Links

Text available via DOI:

View graph of relations

A coherent approach to Bayesian Data Envelopment Analysis

Research output: Contribution to journalJournal articlepeer-review

Published
<mark>Journal publication date</mark>1/03/2020
<mark>Journal</mark>European Journal of Operational Research
Issue number2
Volume281
Number of pages10
Pages (from-to)439-448
Publication StatusPublished
Early online date28/08/19
<mark>Original language</mark>English

Abstract

Mitropoulos et al. (2015) suggested the use of a Bayesian approach in Data Envelopment Analysis (DEA) which can be used to obtain posterior distributions of efficiency scores. In this paper, we avoid their assumption that alternative data sets are simulated from the predictive distribution obtained from their simple data generating process of a normal distribution for the data. The new approach has two significant advantages. First, the posterior proposed in this paper is coherent or principled in the sense that it is consistent with the DEA formulation. Second, and perhaps surprisingly, it is not necessary to solve linear programming problems for each observation in the sample. Bayesian inference is organized around Markov Chain Monte Carlo techniques that can be implemented quite easily. We conduct extensive Monte Carlo experiments to investigate the finite-sample properties of the new approach. We also provide an application to a large U.S banking data set. The sample is an unbalanced panel of US banks with 2,397 bank–year observations for 285 banks. The main purpose of the analysis is to compare distributions of efficiency scores. Relative to DEA, Bayes DEA provides different efficiency scores and their sample distribution has significantly less probability concentration around unity. The comparison with bootstrap-DEA shows that results from Bayes DEA are in broad agreement.

Bibliographic note

This is the author’s version of a work that was accepted for publication in European Journal of Operational Research. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in European Journal of Operational Research, 281, 2, 2020 DOI: 10.1016/j.ejor.2019.08.039