Fuzzy Entropy on Restricted Fuzzy Linear Regression Model with Cross Validation and Applications

The concept of fuzzy regression model with fuzzy regression coefficients was first introduced in [4], [5] and [6]. A fuzzy number can be uniquely determined through its position and entropy as described in [9]. Hence, by using the concept of fuzzy entropy the estimators of the fuzzy regression coefficients may be estimated. In the present communication, we develop a newer fuzzy linear regression (FLR) model with some restrictions in the form of prior information. We have obtained the estimators of regression coefficients with the help of fuzzy entropy for the restricted FLR model. Applications on some hypothetical numerical examples are provided in order to illustrate the proposed model and the obtained estimators. Cross validation of the regression model has been done based on computation of R² and using F-test. The proposed model may find applications in various areas of actuarial sciences, fuzzy time series analysis, management decision making etc.