Simulation of Continuous Qualitative Variables in EconometricModels Using Fuzzy Functions and Numbers

The present paper deals with explanatory dummy variables in econometrics models in two different forms namely, discrete and continuous. In other words we will concentrate not only on a usual binary zero and one dummy variables but also using a fuzzy system to include the case in which dummy variable may appear in a non-discrete form. After introducing a fuzzy system, the application of this system in continuous dummy variables in regression analysis has been investigated. The results indicate that using continuous dummy variables in a fuzzy system version that comparing to the ordinary classic usage of a binary dummy variable we will have a better performance on F and t statistics as well as D-W and R-squares. In other words, using continuous dummy variables in a fuzzy system version will increase the level of significance as well as the explanatory power of the model