A Comparative Study of Seasonal Interval Models for Industrial Time Series Forecasting

In recent years, various seasonal time series models have been proposed for industrial and financial markets forecasting. In each case, the accuracy of time series forecastingare fundamental to make decision and hence the research for improving the effectiveness of seasonal forecasting models havebeen curried on. Many researchers have compared different seasonal time series models together in order to determine moreefficient once in industrial and financial markets. In this paper, performance of four seasonal interval time series models including seasonal autoregressive integrated moving average (SARIMA), fuzzy seasonal autoregressive integrated moving average (FSARIMA), fuzzy seasonal multi-layer perceptron(FSMLP), and Watada models are compared together. Empirical results indicate that the FSMLP model is more satisfactory thanother those models. Therefore, it can be a suitable alternativemodel for seasonal interval forecasting of industrial and financial time series.