Hybrid adaptive fuzzy time series model to forecast TAIEX

Over the past few years, fuzzy time series model has been widely researched. However, previous studies have a problem that determines subjectively the length of intervals. Furthermore, the consideration of a forecasting stage only discusses the relations for previous period and next period. This paper propose a promising hybrid model to get more efficient forecasting. Hence, this study proposes a fusion model, which incorporates a granular spread partition method and the adaptive expectation method to enhance the forecasting results. To verify the proposed model, a ten-year period of the TAIEX (Taiwan Stock Exchange Capitalization Weighted Stock Index) are employed as experimental datasets. From the experiment results, the performances of proposed integrated model surpass the listing models.