Seasonal ARIMA for Forecasting Sea Surface Temperature of the North Zone of the Bay of Bengal

Abstract

The behavior of the sea surface temperature (SST) of the north zone of the Bay of Bengal plays an important role for understanding climate changes over Bangladesh. The monthly average of SST of this zone is used in this study which is obtained from January 1900 to December 2009. Box and Jenkins method is used to fit a seasonal auto regressive integrated moving average (ARIMA) model used in forecasting SST of the north zone of the Bay of Bengal. The most commonly used model selection criteria’s such as the Akaike’s information criterion (AIC), the Bayesian information criterion (BIC), etc. are used for model comparison. The Root Mean Squared Error, Mean Absolute Error and Mean Absolute Percent Error are also used for diagnostic checking in model selection procedure. Seasonal ARIMA (2, 0, 1) (0, 1, 1)12 model is suggested for forecasting the SST of the north zone of the Bay of Bengal.

Keyword: Sea surface temperature, seasonal ARIMA, Akaike’s information criterion, Bayesian information criterion, forecasting value