ABSTRACT
The importance of ocean shipping for international trade forecasting is growing due to the specialization and evolution of industrial sectors around the world. Classic approaches for cargo volume forecasting, such as time series and casual methods, may have poor performances if the data size is small with large fluctuations. This study proposes a hybrid forecasting model based on the Grey forecasting models and an industry share transformation technique. The hybrid model is particular useful for problems when there are dynamic changes in the industry share and the sample size in historical dataset is small. Using a case study of cargo export and import by industry between Taiwan and North American, the proposed model shows good forecasting performances. The findings can be useful for the marine carriers in responding to the dynamic industrial changes.
Highlights
We develop a hybrid model based on Grey theory and share forecast for the cargo volume forecasting problem.
A share forecast component is used to smooth the trend and capture the industrial changes.
A Markov-Chain residual sign estimation is used to modify abnormal fluctuation in the trend.
A case study using import and export cargo by industry between Taiwan and North America is used to demonstrate the performance of the model.
The hybrid model outperforms the other forecasting models compared.
Acknowledgments
The authors grateful acknowledge the two reviewers for their valuable comments on an earlier version of the paper.
Disclosure statement
No potential conflict of interest was reported by the authors.