Modelling of Banking Crisis Forecasting in Iran by BMA

Banking crises are occurring intermittently, indicating that predictive warning models are currently unsuccessful in identifying these crises before they occur. Examining the existing models shows that the reason for the failure of these models is mainly due to the identification of explanatory variables and the experimental design of the model, which were tried to be improved in this research. In order to adjust the model uncertainty problem, this research has determined the effective factors on banking crises in Iran by averaging all the models (Bayesian averaging). The results show that among the BMA, TVP-DMA and TVP-DMS, BVAR and OLS models, the TVP-DMA model was determined as the most efficient model. Based on the model, 10 fragile variables affecting the banking crisis were identified. Based on the results, all the variables have a positive effect on the banking crisis, and this shows the unfavorable banking situation, and the banking crisis index in Iran's economy is a problem with wide dimensions; Because variables related to monetary and financial policies affect this index