Eliciting interval probabilities via interval-valued pair-wise comparisons

In recent years, dealing with uncertainty using interval probabilities, such as combination, marginalization, condition, Bayesian inferences, is receiving considerable attention by researchers. However, how to elicit interval probabilities from subjective judgment is a basic problem for the applications of interval probabilities. In this paper, interval-valued pair-wise comparison of possible outcomes is considered to know which one is more likely to occur. Quadratic programming based model is proposed for estimating interval probabilities. Expectation and decision criteria under interval probabilities are given.