About This Event
This webinar will present two papers from the CAS Reserve Call Paper program.
Both deal with ways to include correlation in estimating reserve ranges.
“Making Reserve Ranges More Realistic” by Clark, Ding and Zhou shows how correlation can be included in bootstrapping via a copula.
Backtesting suggests that bootstrapping approaches lead to reserve ranges that may be too narrow. To improve the current methods, we relax England and Verrall’s assumption that each cell in the incremental development triangle is independent by introducing a correlation structure in the resampling of the bootstrap. Backtesting with
Schedule P data suggests that this approach yields reserve ranges that are more reasonable.
“Quantifying Reserve Risk Based on Volatility in Triangles of Estimated Ultimate Losses” by Feng and Robbin examines correlations in age-to-age factors, and makes use of past estimates of ultimate loss.
