Application of Fuzzy Logistic Regression in Modeling Food Patterns

Application of Fuzzy Logistic Regression in Modeling Food Patterns

Nutrition- School of Nutrition and Food of Medical Sciences-Isfahan- Iran. Abstract. In the logistic regression analy sometimes, the observations of the binary response variable are imp e and due to lack of suitable instruments or well defined criteria, we are not able to define the state of binary response variable precisely. Because of the vague nature of the response variable in this situation, a probability distribution Can’t be considered for it. So, the probabilistie: assumptions of the logistie: model is not fulfilled. In some other situations, the relationship between variables is not precise enough to be modeled by ordinary logistic regression. In these situations, the fuzzy logistic regression would be a suitable alternative choice. In this paper, we investigate fuzzy logistic regression and using a real data set, we explain its application by a numerical example in a dietary pattern study.