A mathematical-programming approach to fuzzy linear regression analysis

Most of previous studies on fuzzy regression analysis have a common characteristic of increasing spreads for the estimated fuzzy responses as the independent variable increases its magnitude, which is not suitable for general cases. In this paper, fuzzy linear regression models with fuzzy/crisp output, fuzzy/crisp input are considered, and an estimated method along with a mathematical-programming-based approach is proposed. The advantages of the proposed approach are simplicity in programming and computation, and minimum difference of total spread between observed and estimated values. (C) 2003 Elsevier Inc. All rights reserved.