A method for integrating information obtained from loss data and scenario analysis is presented in this paper. The stochastic process that generates losses within a unit of measure is modeled as a superposition of various subprocesses that characterize individual "loss-generating mechanisms" (LGMs). An end-to-end method is provided for identifying LGMs, performing scenario analysis and combining the outcomes with relevant historical loss data to compute an aggregate loss distribution for the unit of measure. It is shown how the preferred output of scenario analysis can be straightforwardly encoded into a nonparametric Bayesian framework for integration with historical loss data.
Back to Top