de Travail du Centre d ’ Economie de la Sorbonne Operational risk : A Basel II + + step before Basel
نویسندگان
چکیده
Following the Banking Committee on Banking Supervision, operational risk quantification is based on the Basel matrix which enables sorting incidents. In this paper we analyze these incidents in depth and propose strategies for carrying out the supervisory guidelines proposed by the regulators. The objectives are as follows: • On the first hand, banks need to provide a univariate capital charge for each cell of the Basel matrix. That requires constructing Loss Distribution Functions (LDFs), which implies estimating a frequency and a severity distribution. We show that the choice of the theoretical distributions to build the LDFs has a tremendous impact on the capital charges, especially if we do not take into account extreme losses. • On the other hand, banks also need to provide a global capital charge corresponding to the whole matrix. We highlight that a lack of consideration or a poor appreciation of the dependence structure may lead to incorrect capital charges. • Finally, we draw the attention of regulators and managers to two crucial points: 1. The necessity of splitting information sets in two parts while adjusting the severity distribution. The first covering small and medium losses, and the latter containing extreme losses (this point implies problems of granularity mentioned in the last Basel II guidelines), 2. The choice of the risk measure which provides the capital amount. We emphasize that the expected shortfall measure enables a better anticipation of large incidents pertaining to operational risks.
منابع مشابه
de Travail du Centre d ’ Economie de la Sorbonne An efficient threshold choice for operational risk capital computation
Operational risk quantification requires dealing with data sets which often present extreme values which have a tremendous impact on capital computations (VaR). In order to take into account these effects we use extreme value distributions, and propose a two pattern model to characterize loss distribution functions associated to operational risks: a lognormal on the corpus of the severity distr...
متن کاملde Travail du Centre d ’ Economie de la Sorbonne Decision theory under uncertainty
We review recent advances in the field of decision making under uncertainty or ambiguity.
متن کاملde Travail du Centre d ’ Economie de la Sorbonne Guilbaud ’ s Theorem : An early contribution to judgment aggregation Daniel ECKERT , Bernard MONJARDET 2009
In a paper published in 1952 the French mathematician Georges-Théodule Guilbaud has generalized Arrows impossibility result to the "logical problem of aggregation", thus anticipating the literature on abstract aggregation theory and judgment aggregation. We reconstruct the proof of Guilbauds theorem, which is also of technical interest, because it can be seen as the rst use of ultra lters in...
متن کاملde Travail du Centre d ’ Economie de la Sorbonne Guilbaud ’ s Theorem : An early contribution to judgment aggregation
In a paper published in 1952 the French mathematician Georges-Théodule Guilbaud has generalized Arrows impossibility result to the "logical problem of aggregation", thus anticipating the literature on abstract aggregation theory and judgment aggregation. We reconstruct the proof of Guilbauds theorem, which is also of technical interest, because it can be seen as the rst use of ultra lters in...
متن کاملde Travail du Centre d ’ Economie de la Sorbonne On the value of randomization
An optimal contract may involve randomization when the agents differ in their attitudes towards risk, so that randomization enables the principal to relax the incentive constraints. The paper provides a necessary and sufficient condition for local random deviations to be welfare improving in a neighborhood of a nonrandom optimum. JEL classification numbers: H21, H23, H26.
متن کامل