Further evidence against independence preservation in expert judgement synthesis (Q1089692)

The authors present arguments against the following recommendation to a decision maker: form your own probability distribution by combining the opinions of a number of experts in such a way as to preserve any form of expert agreement regarding the indepence of the events of interest. It is shown that for probability spaces containing at least five points a large class of seemingly reasonable combination methods excludes all independence preserving formulas except those which pick a single expert. In the case where at most four alternatives are present, the same conditions admit a richer variety of non-dictatorial methods which are also characterized.