Quantica Risk Modelling
Reserving under uncertainty: ranges beat point estimates
A single best estimate hides the information the board most needs.
By Jonas Mohamed Osman Abdelghafour, known as Yonas Osman (Yonas Osman) ·
Reserving methods differ in what they assume about stability. Chain ladder assumes development patterns persist; Bornhuetter-Ferguson leans on an a priori loss ratio; frequency-severity methods separate the drivers. Choosing among them is a judgement, and judgement should be disclosed.
Stochastic reserving adds a distribution around the estimate. Its value is not the percentile itself but the conversation it starts about which segments carry the most uncertainty and why.
Quantica documents expert judgement as a named, justified input rather than letting it disappear into a selected development factor.
Jonas Mohamed Osman Abdelghafour, known as Yonas Osman, is an actuary and FRM who has spent more than twenty years building and validating risk models across banking, insurance and marine markets.
Key takeaways
- Method choice is a disclosed judgement.
- Stochastic ranges surface where uncertainty concentrates.
- Never hide judgement inside a selected factor.
Author bio
Jonas Mohamed Osman Abdelghafour, known as Yonas Osman is an actuary, FRM and financial risk professional specialising in banking, insurance, model risk, capital modelling and quantitative risk management.