Application of Bayesian Inference in the Structural Modelling of Operational Risk of Chinese Commercial Bank

With the increase of the financial crimes in Chinese commercial banks, to measure the operational risks and build the comprehensive framework of operational risk management become increasingly urgent. In present study, Bayesian inference, which combines the expert opinions and the dynamic risk factors, has been applied to the measurement of the operational risk based on its construction, initiation, parametric learning and scenario analysis. The internal and external frauds were characterized by the Pareto distribution, while the business disruption and the system failures were fitted by logarithm normal model in the process of initiation and elicitation. The results of the scenario analysis showed that the operational risk could be discovered in time and controlled by enhancing the construction of the incentive mechanism and consummating the audit system.