Application of fuzzy time series models for forecasting the amount of Taiwan export

The study proposed Traditional Time Series Method (ARIMA model and Vector ARMA model) and Fuzzy Time Series Method (Two-factor model, Heuristic model, and Markov model) for the forecasting problem. The real world case of Taiwan exports is employed for models' test to compare the forecasting ability among models and to examine the effects of different lengths of interval and increment information on the forecasting error of models. The results indicate that Fuzzy Time Series Method performs better forecasting ability in short-term period prediction, especially Heuristic model. The ARIMA model generates smaller forecasting errors in longer experiment time period. Nevertheless, introducing increment information is not necessarily in improving the forecasting ability of fuzzy time series. As a result, it is more convenient to use the fuzzy time series method in the limited information and urgent decision-making circumstance.