FORECASTING OF MONTHLY RAINFALL IN CUMILLA OF BANGLADESH, USING SARIMA MODEL

In this paper, seasonal autoregressive integrated moving average (SARIMA) model is developed to predict monthly rainfall in Cumilla using data for the period 1971 to 2017.The stationary condition of the data series are observed by ACF and PACF plots and then checked using the statistic such as Augmented Dickey-Fuller Test (ADF). The ADF test confirms that the monthly rainfall is stationary because the p-value of 0.01 is less than 0.05.The model for which the values of the criteria are smallest is considered as the best model. We found that the SARIMA (1,0,0)(2,1,0)[12]has been fitted to the data based on the Akaike Information Criterion (AIC), Corrected Akaike Information Criterion (AIC_C) and Schwarz Bayesian Information Criterion (BIC). Using Root Mean Square Error (RMSE), Absolute Mean Error (AME) and Mean Absolute Percentage Error (MAPE) to measure forecast accuracy. Then forecast of the data have been made using selected type of SARIMA model for the next five years.