Subjective multi-prior probability: A representation of a partial likelihood relation

This paper deals with an incomplete relation over events. Such a relation naturally arises when likelihood estimations are required within environments that involve ambiguity, and in situations which engage multiple assessments and disagreement among individuals' beliefs. The paper characterizes binary relations over events, interpreted as likelihood relations, that can be represented by a unanimity rule applied to a set of prior probabilities. According to this representation an event is at least as likely as another if and only if there is a consensus among all the priors that this is indeed the case. A key axiom employed is a cancellation condition, which is a simple extension of similar conditions that appear in the literature.