Towards improved decision support in financial sector operational risk analysis
Kokkonen, Mika (2023)
Pro gradu -tutkielma
2023
School of Business and Management, Kauppatieteet
Kaikki oikeudet pidätetään.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi-fe20231030141810
https://urn.fi/URN:NBN:fi-fe20231030141810
Tiivistelmä
As all organizations seek to achieve their objectives, they make decisions under uncertainty and attempt to understand what might happen, both on positive and negative, as much as possible for their benefit. Risk management is part of the uncertainty identification and reduction, thus being an important part of decision-making. Thus, risk management should provide useful support to decisions to create value in taking actions to achieve objectives.
However, over last few decades risk management has suffered from oversimplification to make it accessible to wider audience. This has led to models that have been developed outside of mathematics and decision analysis, and thus they are not considered useful or improving decision-making under uncertainty. Operational risk is one part of wider risk management and should also provide useful insight into decision making with information created via useful models. However, unlike other financial sector’s risk classes like credit risk, operational risk does not have consensus how to build models for its analysis.
The purpose of this thesis was to assess prevalent model of operational risk analysis and how modelling approach should be improved to support decision making. Via design science re-search, qualitative and quantitative modelling of the decision situation was done with influence diagrams, bowties, and Monte Carlo simulation. Related risks were quantified with Monte Carlo simulation to present their uncertain effect on the decision objective. The study shows that applied methods are better in operational risk analysis than prevalently used risk matrix. Another contribution is that it was shown that risk matrix is not risk analysis, but classification. Lastly, it was shown that there seems to be no consensus on what operational risk analysis model should consist of. All in all, the study contributed to financial sector operational risk analysis by providing a practical blueprint with steps towards better decision support with useful models.
However, over last few decades risk management has suffered from oversimplification to make it accessible to wider audience. This has led to models that have been developed outside of mathematics and decision analysis, and thus they are not considered useful or improving decision-making under uncertainty. Operational risk is one part of wider risk management and should also provide useful insight into decision making with information created via useful models. However, unlike other financial sector’s risk classes like credit risk, operational risk does not have consensus how to build models for its analysis.
The purpose of this thesis was to assess prevalent model of operational risk analysis and how modelling approach should be improved to support decision making. Via design science re-search, qualitative and quantitative modelling of the decision situation was done with influence diagrams, bowties, and Monte Carlo simulation. Related risks were quantified with Monte Carlo simulation to present their uncertain effect on the decision objective. The study shows that applied methods are better in operational risk analysis than prevalently used risk matrix. Another contribution is that it was shown that risk matrix is not risk analysis, but classification. Lastly, it was shown that there seems to be no consensus on what operational risk analysis model should consist of. All in all, the study contributed to financial sector operational risk analysis by providing a practical blueprint with steps towards better decision support with useful models.