Conformal prediction for frequency-severity modeling
This work addresses uncertainty quantification in insurance risk modeling, providing a practical tool for actuaries and insurers, though it is incremental as it adapts existing conformal prediction methods to a specific domain.
The authors tackled the problem of constructing prediction intervals for insurance claims with finite sample guarantees by extending split conformal prediction to two-stage frequency-severity modeling. They demonstrated the framework's effectiveness on simulated and real datasets, achieving results with statistical guarantees and leveraging out-of-bag mechanisms to eliminate calibration sets in some cases.
We present a model-agnostic framework for the construction of prediction intervals of insurance claims, with finite sample statistical guarantees, extending the technique of split conformal prediction to the domain of two-stage frequency-severity modeling. The framework effectiveness is showcased with simulated and real datasets using classical parametric models and contemporary machine learning methods. When the underlying severity model is a random forest, we extend the two-stage split conformal prediction algorithm, showing how the out-of-bag mechanism can be leveraged to eliminate the need for a calibration set in the conformal procedure.