A Note on Evaluation of Fuzzy Linear Regression Models by Comparing Membership Functions

Kim and Bishu (Fuzzy Sets and Systems 100 (1998) 343-352) proposed a modification of fuzzy linear regression analysis. Their modification is based on a criterion of minimizing the difference of the fuzzy membership values between the observed and estimated fuzzy numbers. We show that their method sometimes does not find acceptable fuzzy linear regression coefficients. To overcome this shortcoming, we modify their method. Finally, two numerical examples are solved in order to illustrate efficiency of the modified method