On fuzzily measurable random variables (Q1114214): Difference between revisions

Previously a fuzzy random variable was defined as a mapping from a probability space into a space of fuzzy sets of \({\mathbb{R}}^ n\) subject to certain measurability requirements. See for instance \textit{R. Féron} [Publ. Econ. 12, No.1, 81-118 (1979; Zbl 0406.46010)] and \textit{M. L. Puri} and \textit{D. A. Ralescu} [J. Math. Anal. Appl. 114, 409-422 (1986; Zbl 0592.60004)]. In this paper the author takes a different point of view: a mapping \(x:\Omega \to {\mathbb{R}}\) is a fuzzy random variable if it is fuzzy measurable with respect to a fuzzy \(\sigma\)-algebra \(\sigma\) on \(\Omega\) and a fuzzy soft \(\sigma\)-algebra \(\beta\) on \({\mathbb{R}}\) generated by a fuzzy relation ``less than''. The author proves that if x is a fuzzy random variable then there exists an ordinary \(\sigma\)-algebra \({\mathcal A}\) on \(\Omega\) such that x is measurable with respect to \({\mathcal A}\). The converse is also true under some additional assumptions. In addition, he defines a cumulative distribution function of a fuzzy random variable and studies its properties.