Forecasting Container Throughput within Multi-Port Region Using CDE-MPR Methodology: the Case of PRD Region

This paper proposes a CDE-MPR methodology designed to forecast and analyze container throughput volumes. The proposed methodology includes: a container throughput module which considers both interactions among ports and other influence factors, a demand demand-oriented module which provides additional information and chance events effect had on the maritime industries. It is applied to forecast container throughput volumes of the ports in the Pearl River Delta (PRD) region of South China after global financial crisis. To evaluate the forecasting performance of CDE-MPR methodology, we compare its performance with those of ARIMA, VEC and VAR. The empirical results show that CDE-MPR method substantially outperforms other single-model methods. It also demonstrates the consequences of the global financial crisis on each port in PRD region and the weakening effects of the financial crisis on each port during the second half of year 2009, especially on Guangdong port. The study for PRD region also confirms the necessity of analyzing and forecasting from a multi-port region perspective. It is also feasible and effective to apply and expand the CDE-MPR methodology to other similar issues.