Fuzzy time-series based on adaptive expectation model for TAIEX forecasting

Time-series models have been used to make predictions in the areas of stock price forecasting, academic enrollment and weather, etc. However, in stock markets, reasonable investors will modify their forecasts based on recent forecasting errors. Therefore, we propose a new fuzzy time-series model which incorporates the adaptive expectation model into forecasting processes to modify forecasting errors. Using actual trading data from Taiwan Stock Index (TAIEX) and, we evaluate the accuracy of the proposed model by comparing our forecasts with those derived from Chen’s [Chen, S. M. (1996). Forecasting enrollments based on fuzzy time-series, Fuzzy Sets and Systems, 81, 311–319] and Yu’s [Yu, Hui-Kuang. (2004). Weighted fuzzy time-series models for TAIEX forecasting. Physica A, 349, 609–624] models. The comparison results indicate that our model surpasses in accuracy those suggested by Chen and Yu.