A Comparative between Fuzzy Regression and Robust Fuzzy Regression

Abstract

The research aims at comparing between fuzzy regression and robust fuzzy regression. These models are used in case of the model suffer from fuzziness but not randomness, or both together (fuzziness and random). The fuzzy regression was used to calculate the membership function, which represents the weight matrix, by which the weight of each observation is determined and subsequently calculate the extent of each contribution to the information and then to be used in the process of determining the parameters. .For the purpose of analysis, environmental variables were adopted as response variable (carbon dioxide) and the economic activities as explanatory variables (GDP, population, energy consumption, economic openness). In order to compare the two methods and to determine the efficiency of the estimation, the mean square percentage error was used. The results indicated that robust fuzzy regression was better in estimating parameters than the fuzzy regression.