With the development of the international finance and the deregulation of government regulation, making changes in the operating environment of domestic financial institutions, the financial transactions become more internationalization and liberalization, types of business operated by domestic banks increasingly complex and diverse, species also increases for the risk. The Basel Committee on Banking Supervision noted that the risks faced by banks, the operational risk accounted for about 30%, just below the credit risk, which indicates that the impact and losses of operational risk events for the banking sector should not be underestimated and must be strict control and careful assessment .
From the literature review, we can see that most studies have concentrated on how to measure operational risk loss based on the New Basel Capital Accord. Meanwhile, studies on the conduction mechanism and the dynamic evolution of operational risk are relatively scarce. Therefore, this study attempts to establish an abstract operational risk conduction model to explore operational risk of the conduction mechanism, conduction process and simulation of conduction result, taking the correlation between the sources of different types of operational risk into account to measure VaR and ES, and compare with the results of extreme value theory model. The empirical results as following:
1.After the scale-free network model considering the average correlation between different types of operational risk events, the results of VaR and ES measured by the scale-free network model are higher than the extreme value theory model.
2.The largest VaR and ES measured by first class losses (internal fraud) after the operational risk classification, therefore, financial institutions must establish a good internal control measures to prevent this type of operational risk events occur.