Fuzzy regression models and alternative operations for economic and social sciences

Abstract

The economic and social research is often based on assumptions that have little to do with reality, because all scientific knowledge of the Western world is based on the Aristotelian binary logic. Often researchers use rigid conventions and treat things as if they were black or white, but the world around us is nuanced, in other words, is fuzzy. In particular, problems inherent the inaccuracy of measurements and the vagueness of linguistic attributes has led many scholars to consider the fuzzy theory during the recent decades. In this paper we face a particular aspect of fuzzy logic, that is, fuzzy regression. After a brief review of the recent studies about this topic, we focus on some limitations of the previous approaches and suggest some possible solutions. Specifically, we propose the use of alternatives operations to solve some problems of addition and multiplication between fuzzy numbers in fuzzy regression models.