Title :
A formula for fuzzy linear regression analysis
Author_Institution :
Dept. of Appl. Math., Nat. Univ. of Tainan, Tainan, Taiwan
Abstract :
The purpose of this paper is to deal with the problem of least-squares multiple regression with fuzzy data. The constant regression coefficient is assumed to be symmetric triangular, and other coefficients are real (crisp). By applying symmetric triangular approximations of fuzzy numbers, a new method for computing the regression coefficients is proposed. The new method is efficient and easy to determine the coefficients.
Keywords :
fuzzy set theory; geometry; least squares approximations; number theory; regression analysis; constant regression coefficient; fuzzy data; fuzzy linear regression analysis; fuzzy numbers; least-squares multiple regression; symmetric triangular approximations; Computational modeling; Diamond-like carbon; Fuzzy sets; Least squares approximation; Linear regression; Mathematical model; fuzzy linear regression; least-squares estimates; symmetric triangular approximation;
Conference_Titel :
Fuzzy Systems (FUZZ), 2011 IEEE International Conference on
Conference_Location :
Taipei
Print_ISBN :
978-1-4244-7315-1
Electronic_ISBN :
1098-7584
DOI :
10.1109/FUZZY.2011.6007563