Random fuzzy sets: why, when, how
-
Autores: María Angeles Gil Alvarez
, Ana María Colubi Cervero , Pedro Nicolás Terán Agraz
- Localización: BEIO, Boletín de Estadística e Investigación Operativa, ISSN 1889-3805, Vol. 30, Nº. 1, 2014, págs. 5-29
- Idioma: inglés
- Enlaces
-
Resumen
Random elements of non-Euclidean spaces have reached the forefront of statistical research with the extension of continuous process monitoring, leading to a lively interest in functional data. A fuzzy set is a generalized set for which membership degrees are identified by a [0, 1]-valued function.
The aim of this review is to present random fuzzy sets (also called fuzzy random variables) as a mathematical formalization of data-generating processes yielding fuzzy data. They will be contextualized as Borel measurable random elements of metric spaces endowed with a special convex cone structure. That allows one to construct notions of distribution, independence, expectation, variance, and so on, which mirror and generalize the literature of random variables and random vectors. The connections and differences between random fuzzy sets and random elements of classical function spaces (functional data) will be underlined. The paper also includes some bibliometric remarks, comments on the statistical analysis of fuzzy data, and pointers to the literature for the interested reader
