Bank efficiency assessment using a hybrid approach of random forests and data envelopment analysis

Abstract

This study introduces a three-stage integrated framework consisting of data envelopment analysis (DEA), random forest, and logistic regression to examine and predict the impact of environmental variables on banks' performance. This framework identified five important environmental variables and their effects on bank performance when applied to 151 banks in Middle East and North African (MENA) countries over the period 2008-2010.