An Evaluation on the Wind Speed Forecasting Model Using the ARIMA and SARIMA Methods Based on the MSE Value

Wind speed in archipelago areas has an impact on fishing activities, transportation, ocean currents and wave height, as well as maritime weather. An accurate information about wind speed is needed. Wind speed forecasting data within a certain time can be used for decision making in transportation as well as the development of renewable energy technologies. Wind speed is a type of time series data. The ARIMA and SARIMA, are methods of evaluating and analyzing time series data that are widely used in this era. This study uses the ARIMA and SARIMA methods to forecast daily wind speed data from October 2022 to March 2023. This data is the result of measurements from the Meteorology, Climatology and Geophysics Agency (BMKG) Hang Nadim Batam station. Model accuracy is an important factor when analyzing time series data. Therefore, the best ARIMA and SARIMA models will be determined based on the smallest MSE value. The four best models were obtained from the stages of the temporal model determination process by observing ACF and PACF graphic patterns, namely ARIMA(2,0,0), ARIMA(0,1,1), ARIMA(1,1,1) and SARIMA(1,1,0)(1,1,1)s. This study finding that the ARIMA (1,1,1) model is the best model with the smallest MSE value 6.72465.