| Along with operational risk includ,ed by the new Captial Accord,how to accurately measure the operational risk of banking industry has become a hot research issue.The technology of operational risk measurement is still in its infancy compared with that of market risk and credit risk. In our commercial banks, the measurement technology of operational risk is seriously lagging behind, especially, large losses of operational risk events occur frequently in recent years, which make the operational risk have almost become"normal" risk in banks,so it is greatly significant to research the problems of operational risk measurement.In this paper,we chose the operational risk loss data in our banking as a research sample and divided loss events into four types:internal fraud,external fraud,illegal operation and system failure. Then, Based on the analysis of loss distribution approach, we apply two-stage distribution to fit the loss intensity distribution of operational risk,Meanwhile,we apply Conjugate prior distribution to determine the posterior distribution of each parameter and then use Gibbs sampling of Bayesian theory to obtain the parameter estimates, which can reduce error caused by the insufficient low-frequency and high-loss data. Finally, we calculate VaR and ES for different confidence level of the single types of operational risk of commercial banks in China. In view of the correlation between different types of operational risk loss, the copula function is applied to integrate the total loss distribution. By the emprical analysis of the result shows that:Parameter estimation based on Bayesian theory takes into account a priori information such as population and sample information which can reduce the estimated error. The introduced copula function and measured value of VaR and ES not only consider the probability of loss events, which can also calculate potential losses of operational risk, so it can get a more accurate measurement result of operational risk. |
PDF Full Text Request