Application of Hybrid Approach in Banking System: An Undesirable Operational Performance Modelling

Abstract

Non-performing loans (NPLs) is a critical constituent that impacts the operational performance of banks. Rising level of risk leads to poor operational performance, especially when it is beyond the bank’s capabilities to control the increasing bad assets. This calls for real-time performance assessment coupled with futuristic decision making to support banking managers. This observation motivates the authors of this article to develop a two-stage performance prediction assessment model. Accordingly, a hybrid approach combining data envelopment analysis (DEA) and artificial neural network (ANN) is developed to measure and predict the operational efficiency scores of banks. DEA effectively explores the operational performance as well as improvable areas of inefficient banks. The training of ANN model is dependent on estimated operational DEA efficiency scores with the objective to estimate the efficiency scores. Domain for the validation of this study includes dataset derived from Indian banks. The validation result shows that trained ANN model has the prediction capacity with minimum error and maximum accuracy. Finally, the outcome of this study is significantly directed towards business managers who can rely on predictions based on empirical findings of this proposed hybrid modelling.