How to combine expert opinions is an issue that has many aspects and even more “answers.” The problem addressed here is the incremental information an additional expert brings when he or she is correlated with the existing experts. R. T. Clemen and R. L. Winkler derive an algebraic formula that gives the surprising answer—usually very little. In this paper, we provide the geometrical intuition behind the Clemen and Winkler result. We also show how the Clemen and Winkler formula breaks down (i.e., gives bizarre results) as the error covariance structure approaches singularity.
No takes yet. Share an insight, caveat, or question.
Morrison et al. (1991) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: