Proposing a Model to Forecast the Efficiency of Bank Branches under Uncertainty Conditions based on SDEA-PCA Approach and Monte Carlo Simulation

Today, the need to use efficiency measurement systems in the modern banking industry has become increasingly apparent. Therefore, the efficiency of banks needs to be forecasted so that future economic growth can be monitored in future decisions. This paper designs a new integrated model based on the Stochastic Data Envelopment Analysis (SDEA) model and the Principal Component Analysis (PCA) method in a dynamic environment to forecast the efficiency of branches in the modern banking industry by considering variable returns to scale for them. In order to deal with the uncertainty in efficiency forecasting, the inputs and outputs of the branches are designed as triangular fuzzy stochastic variables with normal distribution. In this study, Monte Carlo (MC) simulation and meta-heuristic algorithms have been used to solve the proposed model. Finally, in order to evaluate the performance and accuracy of the proposed integrated model, a case study based on modern banking indicators has been presented to forecast the efficiency of the future financial period of the branches and the results have been analyzed.