A hybrid forecast model combining fuzzy time series, linear regression and a new smoothing technique

In recent years, Fuzzy Time Series have been considered a promising tool to deal with forecasting problems due to the ease to model the problems, the satisfactory results obtained and also to the low computational cost required. However, the long experience with traditional methods coming from statistics, certainly brings a rich knowledge that can be used to enhance the computational methods employed to deal with Fuzzy Time Series. This paper introduces a forecast model where Fuzzy Time Series, linear regression and a new smoothing method are combined. Experiments were performed with the Taiwan Stock Exchange index and compared with eight others approaches found in the literature. The results confirm that the proposed model presents a good accuracy with relation to the other methods.