DOI: 10.36347/sjavs.2017.v04i09.004

Pages: 344-349

Fisheries sector plays an important role in Sri Lankan economy. Demand for fish is increasing with time. Therefore, forecasting the fish production is vital for better planning of fisheries development and management practices. This study aimed to find a suitable Autoregressive Integrated Moving Average (ARIMA) model to forecast annual fish production in Sri Lanka. National annual fish production data during the period from 1972 to 2016 were used for the study. During this period, fish production has increased from 101712 MT to 530920 MT with some fluctuations during the period. The minimum annual fish production of 100702 MT was recorded in 1973 while the maximum of 535050 MT has been recorded in 2014. Several ARIMA models were tested and the most appropriate model was selected based on the validity of assumptions and the accuracy of forecasts. ARIMA (1,1,1) model could be selected as the best model and it gave predictions for year 2016 with 1.35% forecasting error. Estimated production for year 2017 is 551046 MT. Developed model would help the decision makers to establish priorities in terms of fisheries management.