(1) Insurance undertakings must have a regular model validation cycle, comprising monitoring the performance of the internal model, reviewing the continuing appropriateness of its specification, and testing the model's results against experience.
(2) The model validation process must include an effective statistical process for validating the internal model, enabling it to be demonstrated to the supervisory authority that the capital requirements calculated using the internal model are appropriate.
(3) The statistical methods applied must test the appropriateness of the probability distribution forecast against observed losses and against all material new data and related information.
(4) The model validation process includes an analysis of the stability of the internal model and, in particular, a review of the sensitivity of the results of the internal model to changes in the key assumptions on which the model is based. It also includes an assessment of the accuracy, completeness, and appropriateness of the data used for the internal model.
Part 2 · Provisions for direct insurance and reinsurance › Division 2 · Solvency requirements › Section 120
Validation standards
←→ also move between sections