LP Methods for Fuzzy Regression and a New Approach


Cetintav B., ÖZDEMİR A. F.

6th International Conference on Soft Methods in Probability and Statistics, Konstanz, Germany, 4 - 06 October 2012, vol.190, pp.183-191 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 190
  • Doi Number: 10.1007/978-3-642-33042-1_20
  • City: Konstanz
  • Country: Germany
  • Page Numbers: pp.183-191
  • Keywords: Fuzzy linear regression, linear programming methods, LINEAR-REGRESSION, OUTLIERS
  • Dokuz Eylül University Affiliated: Yes

Abstract

Linear Programming (LP) methods are commonly used to construct fuzzy linear regression (FLR,) models. Probabilistic Fuzzy Linear Regression (PFLR) [9] and Unrestricted Fuzzy Linear Regression (UFLR) [3] are two of the mostly applied models that employ LP methods. In this study, a modified fuzzy linear regression model which use LP methods is proposed. PFLR, UFLR and proposed model compared in terms of mean squared error (MSE) and total fuzziness by using two simulated and one real data set.