Title :
Research on Operational Risk measurement for commercial bank based on Credal network
Author :
Liang, Li-jun ; Cao, Li-jie ; Li, Zhi-xiang
Author_Institution :
Sch. of Manage. & Econ., Beijing Inst. of Technol., Beijing, China
Abstract :
Credal network is the extension of Bayesian network. It could not only introduce the uncertainty into the probability sets but also integrate various knowledge and opinion of financial experts based on Bayesian network to reason under uncertain circumstance. Credal network method could analyze and get the joint probability interval of the formation nodes of risk factors to make use of the combination of quantitative and qualitative analysis methods. At the same time, this method could establish the Credal network direct acyclic graphs (DAG) of operational risk for commercial banks so as to propose the new reference model for operational risk measurement. The research purpose expects to provide the efficiency and validity methods to apply to the operational risk measurement for commercial banks.
Keywords :
banking; belief networks; decision making; decision theory; econometrics; inference mechanisms; probability; risk analysis; set theory; uncertainty handling; Bayesian network; Credal network; DAG; OR measurement model; commercial bank; decision making; direct acyclic graph; financial expert; joint probability interval set; operational risk factor measurement model; qualitative analysis method; quantitative analysis method; uncertainty reasoning; Banking; Bayesian methods; Business; Conference management; Cybernetics; Machine learning; Performance loss; Risk analysis; Risk management; Technology management; Bayesian Network; Commercial Bank; Credal Network; Operational Risk Measurement;
Conference_Titel :
Machine Learning and Cybernetics, 2009 International Conference on
Conference_Location :
Baoding
Print_ISBN :
978-1-4244-3702-3
Electronic_ISBN :
978-1-4244-3703-0
DOI :
10.1109/ICMLC.2009.5212128