The article presents information on a comparison of two formalisms for managing uncertainty. Any Bayesian treatment of a problem is a belief-function treatment. Suppose the two items of evidence are independent. This means two things: (1) the evidence for Pr sub 1 is independent of the evidence for Pr sub 2, in the sense that these two items of evidence together can be represented by the product probability measure Pr sub 1 X Pr sub 2 on the Cartesian product S sub 1 X S sub 2.
No takes yet. Share an insight, caveat, or question.
Shafer et al. (1990) studied this question.