Time-series modeling of fishery landings in the Colombian Pacific Ocean using an ARIMA model

Abstract

Seer fish (Scomberomorus sierra) and mullet (Mugil cephalus) are some of the most important marine fishery resources along the Colombian Pacific Ocean. The objective of this study was to forecast the landings of seer fish and mullet based on data from time-series annual landings reported by the Food and Agriculture Organization of the United Nations (FAO) from 1971 to 2014. The study considered autoregressive integrated moving-average (ARIMA) processes to forecast the landings of the species. The ARIMA model (5,1,5) for seer fish and ARIMA model (2,2,1) for mullet showed good agreement concerning the observed data on landings based on the Akaike information criterion. The results show the ARIMA model to be a suitable method for analyzing statistics. In data-poor fisheries situations, this method can support potential evaluations of fishery production for decision making and management.