expert: Modeling Without Data Using Expert Opinion

package provides tools to create and ma-nipulate empirical statistical models using expertopinion (or judgment). Here, the latter expressionrefers to a specific body of techniques to elicit the dis-tribution of a random variable when data is scarce orunavailable. Opinions on the quantiles of the distri-bution are sought from experts in the field and aggre-gated into a final estimate. The package supports ag-gregation by means of the Cooke, Mendel–Sheridanand predefined weights models.We do not mean to give a complete introductionto the theory and practice of expert opinion elicita-tion in this paper. However, for the sake of complete-ness and to assist the casual reader, the next sectionsummarizes the main ideas and concepts.It should be noted that we are only interested,here, in the mathematical techniques of expert opin-ion aggregation. Obtaining the opinion from the ex-perts is an entirely different task; seeKadane andWolfson(1998);Kadane and Winkler(1988);Tver-sky and Kahneman(1974) for more information.Moreover, we do not discuss behavioral models (seeOuchi,2004, for an exhaustive review) nor the prob-lems of expert selection, design and conducting of in-terviews. We refer the interested reader toO’Haganet al.(2006) andCooke(1991) for details. Althoughit is extremely important to carefully examine theseconsiderations if expert opinion is to be useful, weassume that these questions have been solved previ-ously. The package takes the opinion of experts as aninput that we take here as available.The other main section presents the features ofversion 1.0-0 of package