Prof Helgard Raubenheimer, director of the Centre for Business Mathematics and Informatics (CBMI) in the Faculty of Natural and Agricultural Sciences at the North-West University (NWU), delivered a presentation demonstrating his expertise to the National Institute for Theoretical and Computational Science (NITheCS) community.
The title of his presentation was “Extremes in risk management – a non-parametric approach to the estimation of the quantiles of compound distributions”. In his opening remarks, Prof Helgard gave an overview of operational risk, which is defined as the risk of loss resulting from inadequate internal processes or external events. He highlighted how operational risk differs from other types of financial risks, emphasising that it has no upside potential and is primarily concerned with losses.
In the same vein, Prof Helgard examined the limitations of traditional parametric methods such as the single-loss and perturbative approximations, which depend on predicting more extreme quantiles of an underlying severity distribution. He argued that inaccurate evaluations of these quantiles may result from faulty parametric distributional assumptions.
To get around this issue, Prof Helgard presented a new non-parametric multiplier technique that is based on the single-loss and perturbative approximations as well as the extreme value theory. This novel approach estimates a less extreme lower quantile of the severity distribution and seeks to increase accuracy without making significant parametric assumptions.
The Monte Carlo simulation, by which the novel approach was assessed, validates Prof Helgard’s innovative non-parametric technique as a reliable alternative for extreme quantile estimation – especially in cases where there is little data available.
Further, Prof Raubenheimer emphasised the need for accurate extreme quantile estimation, stating that incorrect assumptions about parametric distributions can lead to inaccuracies in quantile estimation, ultimately impacting financial stability.
His presentation featured a discussion of the practical implications of using the proposed methods in operational risk management in financial institutions.
Prof Helgard concluded by highlighting the importance of not just historical data, but also expert judgement and scenario analysis in improving risk assessments.
About Prof Helgard Raubenheimer
Prof Helgard holds a PhD in Risk Analysis, which he obtained at the NWU in 2010, and his primary research interest is quantitative risk management. From 2016 to 2021 Prof Raubenheimer served as the head of the Scientific Analytical Services (SAS) laboratory. His responsibilities included organising industry training initiatives and promoting interactions with the SAS Global Academic Programme, both domestically and abroad.
Prof Raubenheimer co-authored a number of peer-reviewed publications that Risk.net, the Actuarial Society of South Africa (ASSA), and the South African Statical Association (SASA) have praised both domestically and internationally.
Prof Helgard Raubenheimer
