The subject of this paper is a one-method-¯ts-all approach to quantify and predict future
losses in insurance. This method is based on a semiparametric estimator which is corrected
by some nonparametric smoothing techniques. A number of alternative kernel functions are
considered for removing boundary bias, resulting from transforming data with a parametric
function to bounded support. We also analyse the crucial point of bandwidth selection in
nonparametric statistics and discuss three di?erent bandwidth methods, where two are new to
the ¯eld. An extensive simulation study is presented between totally eighteen di?erent kernel
density estimators, and we also consider a practical application based on operational risk data.
Operational risk itself is de¯ned as the risk of loss arising from inadequate or failed internal
processes, people and systems or from external even