RANDOMIZING EFFICIENCY SCORES IN DEA USING BETA DISTRIBUTION UNDER ‘UNIFORM SAMPLING’ OF ASSURANCE REGIONS

摘要

A criticism leveled against Data Envelopment Analysis (DEA) is that it is incapable of handlinginput/output data contaminated with random errors, and therefore, efficiency scores reported byDEA may not be realistic. In the past, several researchers have addressed this issue by proposingStochastic DEA models, where they treated inputs and outputs as random variables that followprobability distributions. In this paper, we propose a method to randomize efficiency scores bytreating each score as an ‘order statistic’ of an underlying Beta distribution. We use Thompson etal.’s (1996) DEA model appended with Assurance Regions (AR) randomized by our ‘uniformsampling’. In an application to a set of banks, we demonstrate this method and derive somestatistical results.