Operational Risk Measurement Of Commercial Banks Based On Pair-Copula Function

With some operational risk loss events that have occurred, how to accurately and effectively measure operational risk has become the focus of the majority of scholars and banking concern. In China, the measure of operational risk remains at the exploratory stage, but most of them belong to the qualitative analysis. Therefore, to seek an effective method to measure the operational risk has a great significance.This paper uses the lost data of commercial banks as a sample, and it makes use of the Bootstrap sampling methods to get lost in the gap on the basis of the loss distribution approach. It takes the existence of a certain correlation of the various types of operational risk loss events into account. And it uses Copula theory to integrate the four types of operational risk loss. Because of the multivariate random variables, The Copula function are not the same. Based on it, in this paper, we propose a method of Pair-Copula. It can not only choo se different Copula function, but also more accurately depict the correlation of random variables. It also simplifies the parameter estimation process. On the basis of this, it makes use of the Monte Carlo simulation method to calculate a single class action risk value at risk VaR, Finally, it calculates the overall loss of four types of operational risk. Research indicates; Bootstrap sampling method effectively overcomes the problem of the lack of operational risk data. Based on the model construction of the Pair-Copula method. It can effectively portray the correlation of the variable loss. The introduction of VaR can accurately measure operational risk.