Page 9 - Azerbaijan State University of Economics

Aimene Farid, Bahi Nawel:Operational Risk Estimation Using the Value-at-Risk (VAR)
Method: Case Study of the External Bank of Algeria (EBA)

According to this method, the Bank relies on its statistical data based on its previous
qualitative experience (such as regular reports, periodic review, ...) and quantity (validity
of the measurement method, detailed data on internal and external losses, dates of their
occurrence, and the region or country in which the losses occurred,...), then it depends on
statistical modeling, after measuring the magnitude of these risks using one of the models
(Babayev, 2016) , the special funds necessary to cover them are determined (Jan Lubbe,
2010). There are several methods that fall within the framework of the advanced
measurement approach: the Internal Measure Approach, the Loss Distribution Approach
and the Scorecards method. (mesdaa, 2017).

The Concept of Operational Value Exposure (OpVaR)
A common way to model operational risk is to use an actuarial approach that expresses
the amount of maximum loss expected over a time horizon typically estimated at one
year and at a specified confidence level, this loss figure is called Operational Value at
Risk (OpVaR). (Wong C.Y., 2013).

Stages of Operational Risk Assessment Using Value at Risk
The process of assessing operational risk using VaR goes through the following stages:

Building The Database:
The lack of accurate data on operational risk events is the biggest obstacle to
implementing accurate models to measure operational risk. (Linda, Jacob, & Anthony,
2004) According to Basel II, the main sets of data to be used are: internal loss
database, external loss database, scenario analysis, factors affecting the business
environment and internal control systems:

• Building an internal loss database: Although internal data is most useful in determining
the allocation of operational losses to the bank, it should not be the only data source to
measure operational risk. In fact, internal data may provide information whose quality is
entirely under the control of the bank (as opposed to external databases), can be used as a
check for internal self-assessment, and can allow at least certain types of operational events
to capture the best risk trends and impact of internal risk reduction efforts. (Saita, 2007).

• Building an external loss database: Developing an internal historical database of
operational risk events is very costly and time-consuming, and therefore internal data
must be complemented by external data obtained from other organizations. However,
external data must be measured and adjusted to reflect institutional differences in the
mix of business units, level of activity, geography and risk control mechanisms across
companies. Moreover, competing companies are reluctant to disclose sensitive and
detailed information about their internal processes and procedures to competitors.
(Linda, Jacob, & Anthony, 2004).

9