Comparison of some time series models for the analysis of fisheries data

A 12-year record of monthly average catch per day fished of rock lobster, Jasus edwardsii, from New Zealand was analysed using three forecasting procedures. The three procedures, monthly averages, harmonic regression analysis and auto-regressive integrated moving average (ARIMA) models are briefly described and evaluated using the variances of the residuals of the original observations and forecasts compared with actual data not used in developing the models. An ARIMA model was found to be most suitable in terms of producing forecasts up to 12 months ahead.