an integrated neural networks and mcmc model to predicting bank’s efficiency

In the banking industry, there is intense competition between banks to attract resources and facilities. with the development of new services, bank managers try to improve their services and attract more customer deposits by differentiating between competitors' services. this research uses a two-stage topsis method with the combination of neural network model and monte carlo simulation trading method to analyze and compare bank productivity forecasts with the 4 efficiency criteria of the banking industry. topsis was first used in two steps to rate the efficiency of banks and then a model was created for banking performance with clear forecasting ability. secondly, an mcmc sampling method and ann training were presented. integrated neural networks and mcmcs were used which are consistent with topsis results. the simulation effect of the selected variables was predicted and their effect on performance was observed. the proposed method was used successfully for predicting performance and ranking banks based on the relative importance of performance criteria expressed by considering the performance levels in the topsis method. then, the artificial neural network was modeled using the results obtained from the topsis method, an effective model for appropriate prediction of bank performance. based on the results of the proposed model and the level of importance of performance measures, cost and revenue structure were considered to be the main causes of inefficiency.