Stochastic error, parameter error and model error

Document description

The subject of error and uncertainty is and always has been a key issue in General Insurance. There is a growing interest in complex models of the claims process and computer models of both assets and liabilities. However, it is important that the degree of error in such models is recognised. The true uncertainty about the future outcome exceeds greatly that caused purely by random chance, due to errors in the modelling process. It is important that actuaries both communicate and, where possible, quantify such possible errors. The purpose of this workshop is to discuss some of the issues which may affect the results of a modelling process and to discuss what practical approaches exist to describe the components of uncertainty.