Modhej D, Dahimavi A. Forecasting Bank Branch Efficiency Using by Data Envelopment Analysis and a Cerebellar Model Articulation Controller Neural Network. , J. Mon. Ec.. 2026; 21 (4)
URL:
http://jme.mbri.ac.ir/article-1-752-en.html
1- Department of Mathematics, So.C., Islamic Azad University, Sousangerd, Iran
2- Postdoctoral graduate of Water Resources, Khozestan Water and Power Organization
Abstract: (937 Views)
The banking industry plays a major role in a country's economic growth and development. Thus, the evaluation of bank performance is an important concern for bank managers. Data Envelopment Analysis (DEA) is a pioneering approach in the literature for the assessment of organizations’ efficiency. However, conventional DEA models evaluate efficiency based on observed input and output data and do not directly provide forecasts of future efficiency. To address this limitation, this study proposes a complementary approach that combines DEA, a method for measuring relative efficiency, with an artificial neural network (ANN) for predictive modeling. The Cerebellar Model Articulation Controller (CMAC) is a neural network that resembles a part of the brain known as cerebellum. The CMAC network, with its simple structure, is capable of estimating nonlinear functions, system modelling and pattern recognition. Meanwhile, the CMAC approach has fast learning convergence and local generalization compared to other networks. The proposed method is applied to a sample of 210 Iranian bank branches. Historical data from April 2022 to March 2025 are used to forecast bank branch efficiency for the period from April 2025 to March 2028. The empirical results revealed that the predicted efficiency using the DEA and CMAC network has a good correlation with that calculated by the DEA, indicating that the predicted efficiency using the proposed approach is a reasonable proxy to demonstrate efficiency for the future.
Type of Study:
Original Research - Theoric |
Subject:
Economics Received: 5 Nov 2025 | Accepted: 26 Aug 2026 | Published: 20 Oct 2026