Analysis and prediction of walleye pollock (Theragra chalcogramma) landings in Korea by time series analysis

Forecasts of monthly landings of walleye pollock, Theragra chalcogramma, in Korea were carried out by the seasonal autoregressive integrated moving average (ARIMA) model. The equation of Box-Cox transformation on the walleye pollock catch data which handles nonstationary variance was Y' = (Y-0.31 - 1)/0.31. The model identification was determined by minimum Akaike information criteria (AIC) to identify the model as an objective criterion. Monthly forecasts of the walleye pollock landings for 1993-1994 were compared with the actual landings. Total observed annual landings of walleye pollock in 1993 and 1994 were 16,610 metric tons and 10,748 metric tons, respectively, while the model predicted 10,459 metric tons and 8203 metric tons (absolute percentage error 37.0% and 23.7%, respectively). The spectrum analysis of the monthly catches of walleye pollock in Korea showed some dominant fluctuations in the periods of 12 months, 4.4 years, 12.6 months and 6 months in descending order. (C) 1998 Elsevier Science B.V. All rights reserved.