nonparametric analysis

A conditional full frontier approach for investigating the Averch-Johnson effect. MPRA Paper 35491, University Library of Munich, Germany.

This paper applies a probabilistic approach in order to develop conditional and unconditional Data Envelopment Analysis (DEA) models for the measurement of sectors’ input oriented technical and scale efficiency levels for a sample of 23 Greek manufacturing sectors. In order to capture the Averch and Johnson effect (A-J effect), we measure sectors’ efficiency levels conditioned on the number of companies competing within the sectors.