Robust Bayesianism: Relation to Evidence Theory

摘要: We are interested in understanding the relationship between Bayesian inference and evidence theory. The concept of a set probability distributions is central both robust analysis some versions Dempster-Shafer’s interpret imprecise probabilities as posteriors obtainable from likelihoods priors, which convex sets that can be considered represented with, e.g., DS-structures. Likelihoods prior combined with place’s parallel composition. natural simple combination operator makes all pairwise combinations elements two representing likelihood. Our proposed unique, it has interesting normative factual properties. compare its behavior other fusion rules, earlier efforts to reconcile rule consistent Fixsen/Mahler’s modified Dempster’s (MDS) rule, but not rule. framework liberal allowing significant uncertainty concepts modeled taken care therefore viable, probably only, unifying structure economically taught alternative solutions modeled, compared explained.