الفهرس | Only 14 pages are availabe for public view |
Abstract This thesis aims to introduce and analyze the new double chain ladder model for claims reserving and its application in general insurance business compared to the deterministic methods (especially chain ladder method). To achieve the thesis objectives, the first chapter introduces the preliminary concepts about claims reserving; the claims reserving process; then introducing deterministic methods, such as chain ladder method and the bornhuetter ferguson method also introducing stochastic models, such as mack{u2019}s model and over-dispersed poisson model which is an application of the generalized linear models. Finally, presenting how to use the bootstrap technique to obtain prediction error estimates for reserves. The second chapter introduces the double chain ladder model (advantages in relation to other methods). Then considering the possibility of integrating the tail factor, also enabling the application of the bootstrap technique and finally presenting the extension with the bornhuetter ferguson method. The third chapter presents a practical application of previous exposed theory, enabling a comparison between the chain ladder method and double chain ladder model. It was concluded that double chain ladder model is appropriate for separating the future claim estimates into an Incurred But Not reported (IBNR) and reported but Not settled (RBNS) claims reserves given that it provides explicit parameterization that has an intuitive appeal to the actuary and also useful for applying or explaining actuarial judgment |