Model Fitting and Forecasting of Annual National Coconut Production in Sri Lanka

Abstract

Coconut plantation plays a significant role in the Sri Lankan economy. This study mainly focuses on modeling and analyzing the annual national coconut production of Sri Lanka. The annual national production data from 1964 to 2015 were used for this study. Time series models were fitted by the Box and Jenkins ARIMA approach. Series were tested for stationary using the unit root test (Augmented Dickey Fuller test). Differencing techniques were applied to transform non-stationary series to stationary series. Model diagnostics were accomplished by the residual analysis. The ARIMA (2,1,0) model was selected as the best fitting model for annual national production of coconut in Sri Lanka based on Akaike Information Criterion and the residual analysis. The prediction has been made for 2020 where the total number of nuts would be 2713 millions. Further, we have noticed a 5% reduction in total nuts produce in 2017 whereas which would be 4% by 2020. The drop in annual coconut production anticipates a higher price for coconut by 2020. Therefore, a policy initiation must be in place to face this situation of the coconut industry in Sri Lanka.