Impact of Outliers in Data Envelopment ‎Analysis‎

Abstract

This paper will examine the relationship between "Data Envelopment Analysis" and a statistical concept ``Outlier". Data envelopment analysis (DEA) is a method for estimating the relative efficiency of decision making units (DMUs) having similar tasks in a production system by multiple inputs to produce multiple ‎outputs.‎ An important issue in statistics is to identify the outliers. In this paper, we attempt to investigate the concept of the outliers determination by data envelopment analysis and assess the manner of decision making units when a sample contains an outlier. We will start by providing a review literature. We will then proceed with our proposed method and discuss the strengths and weaknesses of our method. We will provide some numerical results to demonstrate the applicability of our ‎method.‎