A NEW NOTION OF POSSIBILISTIC COVARIANCE

Possibilistic indicators of fuzzy numbers (expected value, variance, and covariance) are an efficient instrument in the modeling of uncertainty phenomena. Various models of uncertainty phenomena have led to several notions of variance and covariance. In particular, the possibilistic models of risk aversion previously studied by one of the authors imposed a notion of variance of a fuzzy number different from those existing in the literature. In this paper, a new notion of covariance of two fuzzy numbers corresponding to the possibilistic variance mentioned is studied. This possibilistic covariance can be used, e.g. in models of possibilistic risk aversion with many parameters.