DEA-R Model-based Efficiency Evaluation of Japanese Banks

Since the conventional data envelopment analysis (DEA) models are unsuitable for handling ratio data, we develop a novel type of mathematical model for efficiency evaluation and analysis that combines DEA and ratio analysis (termed DEA-R model). DEA is a popular and powerful approach for evaluating the relative efficiency of decision-making units with multiple inputs and outputs. However, owing to the increasing prevalence of ratio data (e.g., operating profit per person) in practical applications, integrating DEA with ratio analysis has become necessary. Thus, we develop a novel DEA-R model that integrates the well-defined range-adjusted measure (RAM) DEA model with ratio analysis. The developed model can handle ratio data and allows the incorporation of expert opinions in the selection of output-to-input pairs. These advantages over the conventional DEA model make the developed model a more flexible and effective approach. To demonstrate the validity and superiority of the developed model, we revisit a case study using a dataset of Japanese banks. The results of this application are discussed and several future research directions are provided.