Forecasting ASEAN countries exchange rates using auto regression model based on triangular fuzzy number

Abstract

Exchange rate forecasting is important to represent the expectation of exchange rates future values. The forecasting task is due to the economic factor and the historical data used to forecast are exposed to uncertainty and observational error during data collection. The existing auto regression model only deals with uncertainty exist in the model, not in the data preparation. Uncertainties may contained in the data input and should be treated during data preparation which is an early stage of forecasting process. To date, only few researches discuss intensely on a fuzzy data preparation. However, data treatment during data preparation is important to reduce model’s error due to uncertainty problem. Hence, this paper presents an approach to construct Triangular Fuzzy Number to handle uncertainty in data during data preparation. As the Triangular Fuzzy Number is often used to represent uncertain information in a form of interval, this study proposed a procedure to construct Triangular Fuzzy Number from single point data. In this study, the Triangular Fuzzy Number is built in a form of symmetric triangular with 1%, 3% and 5% spread value. Autoregressive model is then used to forecast the exchange rate of Association of South East Asian Nation (ASEAN) countries. The result in this study shows that the forecasting exchange rate is significantly important to trace the movement of ASEAN countries exchange rates and beneficial in forecasting planning.