Snapper is one of the economically important fish targeted by fishers in the island of Timor. Domestic and foreign markets for fillet and fresh snapper is considerably large. The export volume in 2021 was 4,172,056 kg, and the total value reached 12,452,211 USD. Along with the increasing snapper production in Kupang City and NTT, it is necessary to forecast the future snapper production. Thus, it can be a reference for policy makers in the region in designing fisheries development. This study aims to produce forecasting model for snapper catches at PPI Oeba Kupang. This research applied Autoregressive Integrated Moving Average (ARIMA) and Seasonal Autoregressive Integrated Moving Average (SARIMA), as a forecasting method. The data of snapper production analyzed in this study consisted of 72 months, started from January 2016 to December 2021. Data were obtained from the UPT of the Department of Marine and Fisheries of NTT Province at PPI Oeba Kupang. Data have a non-stationary pattern to the variance, hence a logarithmic transformation is needed, as well as seasonal differences analysis. However, there is no need for non-seasonal differences since the data are stationary with respect to the mean. The results of the identification of the Autocorrelation function and Partial Autucorrelation function Plots are ARIMA models with seasonal factor period 4, ARIMA (P,D,Q) = (1,0,1), while the SARIMA order (P, D, Q) = (2,1,1)4. Based on parameter testing, verification, examination, and testing of suitable models, the best model obtained was (0,0,1)(2,1,1)4.
Keywords: Autoregressive Integrated Moving Average, focecasting model, Seasonal Autoregressive Integrated Moving Average, Snapper.