Weighted Fuzzy Time Series Forecasting Model

Traditional time series methods fail to forecast the problems with linguistic historical data. An alternative forecasting method such as fuzzy time series is needed to deal with these kinds of problems. This study proposes a fuzzy time series method based on trend variations. In experiments and comparisons, the enrollment at the University of Alabama is adopted to illustrate and verify the proposed method, respectively. This paper utilizes the tracking signal to compares the forecasting accuracy of proposed model with other methods, and the comparison results show that the proposed method has better performance than other methods.