PREDICTING INCIDENT SIZE FROM LIMITED INFORMATION

Predicting the size of low-probability, high-consequence natural disas ters, industrial accidents, and pollutant releases is often difficult due to limitations in the availability of data on rare events and futu re circumstances. When incident data are available, they may be diffic ult to fit with a lognormal distribution. Two Bayesian probability dis tributions for inferring future incident-size probabilities from limit ed, indirect, and subjective information are proposed in this paper. T he distributions are derived from Pareto distributions that are shown to fit data on different incident types and are justified theoreticall y. The derived distributions incorporate both inherent variability and uncertainty due to information limitations. Results were analyzed to determine the amount of data needed to predict incident-size probabili ties in various situations. Information requirements for incident-size prediction using the methods were low, particularly when the populati on distribution had a thick tail. Use of the distributions to predict accumulated oil-spill consequences was demonstrated.