A heuristic time-invariant model for fuzzy time series forecasting

Many forecasting models based on the concepts of fuzzy time series have been proposed in the past decades. These models have been applied to predict enrollments, temperature, crop production and stock index, etc. In this paper, we present a simple heuristic time-invariant fuzzy time series forecasting model, which uses prediction accuracy of model observations to train the trend predictor in the training phase, and uses these trend predictor to generate forecasting values in the testing phase. This model can capture the trends of the time series more accurately and hence improve the forecasting results. The proposed method is applied for forecasting university enrollment of Alabama and the Taiwan Futures Exchange (TAIFEX). It is shown that the proposed model achieves a significant improvement in forecasting accuracy as compared to other fuzzy time series forecasting models.