Modelling reserve risk using the re-reserving approach
In general modelling reserve risk seems to be complicated. Several underwriting years and their development years have to be assessed. A full model may contain many random variables or Diriclet type distributions. Is there an easier option?
Yes, there it is. Let us try a stochastic re-reserving approach. The approach looks at the changes of the loss development due to the new information gathered within one period. The incremental paids or incurreds may differ from what the expected and additionally this leads to different estimates from the re-reserving at the end of the period. The fitting of the calendar year risk can be done using the results of the paper "Uncertainty in the claims development result in the chain ladder method." (M. Wüthrich, et. al., 2009).
The PODRA model in PillarOne.RiskAnalytics offers the option to feed the fitted model to the reserve risk section. Adding premium risks to the model and running the aggregation process we can obtain integrated results for premium and reserve risks.
When optimizing reinsurance structures try out reinsurance contracts on reserves like loss portfolio transfer or adverse development cover.
-- Jörg


Example
(...) The incremental paids or incurreds may differ from the expected and additionally this leads to different estimates from the re-reserving at the end of the period. The fitting of the calendar year risk can be done using the results of the paper "Uncertainty in the claims development result in the chain ladder method." (...)
Markus M.