Example: Measuring operational risk exposure

Algorithmics consider the primary areas of operational risk exposure in an investment
banking example to be measured due to transactions. These areas include reconciliation,
valuation, settlement, compliance, and limits. Algorithmics considers each of
these areas in detail and provide exposure to loss using the value of the transaction
that is unreconciled, the difference in a transaction’s value and an independent
valuation, the value of a transaction that is not in compliance, the cost of carry for a
transaction, and the amount of limit violation a transaction causes. Each of these
areas results in an exposure in home currency of potential loss due to operations.
Expected loss is modeled for each of the areas and used to provide estimates such
as value at loss and loss under scenario Exposures are calculated by transaction for each process and a statistical sequence
of exposures over a time period is generated. The measurement of operational risk is
a statistic on this sequence that includes variability of exposure. These measurements
are used to generate quality control charts and provide the input to the
monitoring for goal-based monitoring (target versus actual) and exception monitoring.