On Horizontal Fuzzy Relations and Hypotheses Testing

Abstract

Making decisions in conditions of uncertainty is one of today’s everyday problems. Various statistical models have found their place in practical applications, however working with imprecise, quite often fuzzy valued, data is still a challenge. In this paper, a new fuzzy relation is presented that enables the comparison of triangular fuzzy numbers along the horizontal axis, and a proposal is given for the generalization of the statistical hypothesis for data in the form of triangular fuzzy numbers. The proposed method allows a fuzzy form for both the registered value of the test statistic and the reference critical value and defines the degree of acceptance of the null hypothesis.