Extended Fuzzy Regression Model with Least Squares Estimation and Residual Evaluation

Neng-Fang Tseng

《International Journal of Information and Management Sciences》 23:4 (2012/12) Pp. 341-357

https://doi.org/10.6186/IJIMS.2012.23.4.1

並列摘要

In this paper we extend the basic configuration of the fuzzy regression model so that boundaries of the membership function are less restricted, as well as the procedure of estimation for the fuzzy coefficient is also proposed. In addition, the residuals in the extended fuzzy regression model are evaluated by fuzzy arithmetic. The mean squared error defined in terms of the fuzzy residuals is used as a criterion for model selection. Finally the empirical study shows that the model selection procedure using MSE can provide an objective decision in the analysis of Taiwan's Monitoring Indicator.

並列關鍵字

Extended fuzzy regression model (EFR model) ； fuzzy coefficient ； least squares method ； fuzzy residuals ； mean squared error (MSE)

參考文獻

Andres, J. de.(2006).Calculating insurance claim reserves with fuzzy regression.Fuzzy Sets and Systems.157,3091-3108.

Andres, J. de.,Terceno, A.(2004).Estimating a fuzzy term structure of interest rates using fuzzy regression techniques.European Journal of Operational Research.154,804-818.

Asadzadeh, S. M.,Azadeh, A.,Ziaeifar, A.(2011).A neuro-fuzzy-regression algorithm for improved prediction of manufacturing lead time with machine breakdowns.Concurrent Engineering.19,269-281.

Azadeh, A.,Saberi, M.,Ghaderi, S. F.,Gitiforouz, A.(2010).Estimating and improving electricity demand function in residential sector with imprecise data by fuzzy regression.International Journal of Mathematics in Operational Research.2,405-423.

Chen, L. Y.,Mujtaba, B. G.(2010).Strategic prediction of the business cycle using the fuzzy regression model: a study of the council of economic planning and development in Taiwan.International Journal of Business Forecasting and Marketing Intelligence.1,217-23.