SOME APPROXIMATIONS USEFUL TO THE USE OF DEPENDENT INFORMATION-SOURCES

Expert judgments have been frequently utilized in probabilistic risk a ssessments. The aggregation methods used have often been very simplist ic and have ignored the critical issues of incorporating experts' bias es and inter-expert dependence. On the other hand, theoretical methods available address some of these issues, but suffer from the fact that the input requirements are too demanding. This paper provides practic al guidance on the use of the versatile Bayesian expert judgment aggre gation model. Guidance is provided by conducting a sensitivity analysi s of the experts' weights and posterior standard deviation on the para meters of the Bayesian aggregation model. Experts' weights and posteri or standard deviation are most sensitive to experts' standard deviatio ns and are not as sensitive to inter-expert dependence. Experts' locat ion biases influence the posterior mean, but not the posterior varianc e and weights. The results obtained are also applicable to the more co mmonly used judgment aggregation procedures, such as simple geometric averaging, because the Bayesian aggregation method encompasses these s imple techniques. The results are illustrated with the help of an exam ple and suggest that the uncertainty is underestimated if expert biase s are ignored.