A modified EMD-ARIMA based on clustering analysis for fishery landing forecasting

Abstract

This paper investigates the ability of a new hybrid forecasting model based on empirical mode decomposition (EMD), cluster analysis and Autoregressive Integrated Moving Average (ARIMA) model to improve the accuracy of fishery landing forecasting. In the first step, the original fishery landing was decomposed into a finite number of Intrinsic Mode Functions (IMFs) and a residual by EMD. The second stage, the cluster analysis was used to reconstruct the IMFs and residual into high frequency, medium frequency and low frequency components, and then every component are modelled using ARIMA model. Finally, these predicted results are aggregated into an ensemble result as final prediction. For illustration and verification purposes of the proposed model, monthly fishery landing record data from East Johor has been used as a cases study. The result shows that the proposed model obtained the best forecasting result compared with ARIMA and EMD-ARIMA models.