On Prediction of Future Insurance Claims When the Model Is Uncertain

Abstract

Predictive modeling is arguably one of the most important tasks actuaries face in their day-to-day work. In practice, actuaries may have a number of reasonable models to consider, all of which will provide different predictions. The most common strategy is to first use some kind of model selection tool to select a “best model,” and then use that model to make predictions. However, there is reason to be concerned about the use of the classical distribution theory to develop predictions because these ignore the selection effect. Since accuracy of predictions is crucial to the insurer’s pricing and solvency, care is needed to develop valid prediction methods. In this paper, we undertake an investigation of the effects of model selection on the validity of classical prediction tools and make some recommendations for practitioners.

Keywords: Bootstrap; Post-Selection Inference; Predictive Distribution; Regression; Variable Selection

JEL Classification: C51; C52; G22