Thesis Type: Postgraduate
Institution Of The Thesis: Gazi University, Fen Bilimleri Enstitüsü, İSTATİSTİK ANABİLİM DALI, Turkey
Approval Date: 2020
Thesis Language: Turkish
Student: AYŞENUR AKIN VARGELOĞLU
Supervisor: HAMZA GAMGAM
Open Archive Collection: AVESIS Open Access Collection
Abstract:There is increasing-interest in fuzzy logic-based methods for analyzing uncertainties in real-life problems. As an alternative of classical regression analysis, fuzzy regression analysis is utilized to model complex systems containing uncertainties. Besides, uncertainties between input and output variables of any given system can be examined in detail using intuitionistic fuzzy regression analysis. In this thesis study, intuitionistic fuzzy regression analysis based on the least squares, its theoretical structure and applications are examined. It is aimed to obtained better estimators in virtue of the regression based on intuitionistic fuzzy sets. Classical, fuzzy and intuitionistic fuzzy regression models were fitted for various datasets in the literature. Furthermore, supplier selection problem of a company was investigated in order to demonstrate the feasibility of applying intuitionistic fuzzy regression analysis in real-life. The estimators of classical, fuzzy and intuitionistic fuzzy regression models were evaluated according to root of mean residual square as the well-known performance criterion. Evaluations show that intuitionistic fuzzy regression models give better results than the others. In addition, validities of intuitionistic fuzzy regression models were investigated by using the cross-validation method.