Constraints on belief functions imposed by fuzzy random variables: Some technical remarks on Romer-Kandel

C. Romer and A. Kandel (1995) investigated a new approach of fuzzy data analysis combining Dempster-Shafer theory and fuzzy set theory. Especially, belief measures are derived from fuzzy random variables but proved incompletely. This paper offers a completion using topological properties induced by the Hausdorff metric which is defined on the space of closed intervals of the real numbers. Moreover little corrections of some other assertions in the paper of Romer-Kandel are presented.