New posterior distributions for the incidence of inefficiency in DEA scores

Abstract

This note derives posterior distributions for the incidence of data envelopment analysis generated inefficiency scores when sampling occurs with replacement or whether the population approaches infinity. The posterior distributions are flexible, tractable, and applicable to a variety of empirical settings.

Highlights

• Data envelopment analysis may generate inaccurate efficiency scores, especially when the sample size is small relative to the number of inputs and outputs

• We derive posterior distributions for the incidence of inefficiency in the presence of this inaccuracy.

• The derived posterior distributions are computationally tractable, especially in situation when the inaccuracy is more pronounced.

Keywords:

• Data envelopment analysis
• misclassification
• incidence of inefficiency
• curse of dimensionality

Additional information

Funding

This article was funded by Scientific and Technological Council of Turkey (TÜBİTAK) (project no. 1059B141300829).