Modelling extremal events: for insurance and finance
Modelling extremal events: for insurance and finance
A Bayesian approach to estimate the marginal loss distributions in operational risk management
Computational Statistics & Data Analysis
| Hi-index | 0.00 |
In this paper, we explore the loss data collection exercise for operational risk in Chinese commercial banks from 1999 to first half of 2006. Firstly, the above data are bootstrapped to analyze the capital allocation for a medium-scaled commercial bank in China. Secondly, for every selected cell, we calibrate two truncated distributions to fit the loss severity, one for 'normal' losses and the other for the 'extreme' losses. Moreover, a more realistic dependence structure - multivariate t copula function is used to measure the relation among the selected cells. In the final, the simulation results suggest that substantial savings can be achieved through measuring the dependence by means of multivariate t copula function than by means of perfect positive dependence.