Recent Financial scandals in the banking industry have caused considerable attention to be focused on operational risk. This is because an analysis of some of these scandals reveals that the underlying causes of these huge Financial losses are due to Operational Risk (OR) and not to credit or market risk, as might initially appeared to be the case. The Foreign Exchange (FX) market has had its fair share of these recent scandals. Two most recent examples of operational risk losses in the FX markets are the National Australia Bank's 227 million USD loss in 2004 and Allied Irish Bank's 750 million USD loss in 2002. These losses have had serious negative impact on the firms' profitability and reputation. Besides scandalous losses in the FX market, trade processing and settlement errors, as well as incorrect settlement, if not controlled, may lead to indirect costs such as compensation payments to counter parties, or to the development of large losses in a firm's portfolio due to managing the wrong position. Operational risk losses in the financial industry usually occur at the business unit level and are due to weak management oversight, weak internal controls or the lack of it, or to breakdown of procedures among others. Operational risk therefore has to be managed at the business unit level.
[...] List of Abbreviations ACI International Financial Markets Association AIB Allied Irish Bank AMA Advanced Measurement Approach BIS Bank for International Settlement BN Bayesian Network CFP Committee for Professionalism ACI CLS Continuously Linked Settlement CPT Conditional Probability Tables CSA Control Self Assessment DAG Directed Acyclic Graph FX Foreign Exchange KRI Key Risk Indicator LDA Loss Distribution Approach MCMC Markov Chain Monte -Carlo Simulation MM Money Market NAB National Australian Bank NDT Node Description Table OpVaR Operational Value at Risk OR Operational Risk OTC Over The Counter SI Settlement Instructions SOX Sarbanes Oxley Act STP Straight Through Process VaR Value at Risk Currencies AUD Australian dollar BRL Brazilian real CAD Canadian dollar CHF Swiss franc CZK Czech koruna DKK Danish krone EUR Euro GBP Great Britain Pound Sterling HKD Hong Kong dollar HUF Hungarian forint JPY Japanese yen MXN Mexican peso NOK Norwegian krone NZD New Zealand dollar PLN Polish zloty RUB Russian ruble SEK Swedish krona SGD Singapore dollar SKK Slovak koruna TRL Turkish Lira USD American dollar ZAR South African rand CHAPTER 1 Introduction and Motivation 1.1 Introduction Recent Financial scandals in the banking industry have caused considerable attention to be focused on operational risk. [...]
[...] A survey on algorithms for real-time Bayesian network infer ence. In the joint AASI-02/KDD-02/UAI-02 workshop on Real-Time Decision Support and Diagnosis Systems O. Hansson and A. Mayer. Heuristic search as evidential reasoning. In Proceedings of the 5th Workshop on Uncertainty in Artificial Intelligence M. Haubenstock. The operational risk management framework. In C. Alexander, editor, Operational risk, regulation, analysis and management, London Pearson Education. D. Heckermann. Bayesian networks for data mining. Data mining and knowledge discovery, 1:79{ D. Heckermann and J.S. [...]
[...] CONSTRUCTION OF A BAYESIAN NETWORK FOR FOREIGN EXCHANGE SETTLEMENT computations: compensations less or equal to 100 Euros are considered "de minimum” and not claimed, an administration cost of 100 Euros is payable for failed trades and in situations where the product of the trade value and the days delay is greater or equal to 200 million Euros, the compensation to be paid can be negotiated between the parties concerned. The first provision is catered for in the network by cutting off losses less than 100 euros and normalizing the loss distributions. [...]
[...] An outlook of the prospects and criticism of BN in the context of OR management is given below Prospects of Bayesian Networks for Operational Risk As a tool for managing OR at the business unit level, BN enjoys several advantages over other models. These include the following: 1. A BN is able to incorporate all the three essential elements of the AMA outlined earlier (internal data, external data, scenario analysis, and factors reflecting the business environment and control systems) into one simple model that is easy to understand. [...]
[...] It is hoped that the practical guidance provided in this work will contribute to position Bayesian networks as the preferred tool for managing Operational Risk at the micro (business unit) level, particularly, the managing of operational risk within Foreign Exchange and Money Market settlement process, which has experience huge financial losses recently. The thesis is also intended to challenge the view held by some operational risk practitioners that Bayesian networks are too complex to maintain, too difficult to construct and give little return for the effort put in. [...]
APA Style reference
For your bibliographyOnline reading
with our online readerContent validated
by our reading committee