A Forecasting Model of Dry Bulk Freight Rate Index with Raw Materials Prices

• Main Authors: Yi-jung Ya, 葉一中
• Other Authors: Ching-chih Chang
• Format: Others
• Language: en_US
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/30588736471640287283

碩士 === 國立成功大學 === 交通管理學系碩博士班 === 97 === In recent years, tramp shipping has had a high level of volatility in the dry bulk market. From February 2007 to November 2008, the Baltic Dry Index (BDI) rose 161 percent from 4219 to a historical high of 11039. But the index dropped to 891 in November 2008, record a loss of 92% in a year. Over the past few years due to the global economic growth, most commodities have presented shortage. Therefore, their prices are also reaching their peak. As a result of excess demand, the freight rate and raw materials prices has a high level of volatility. Hence this paper would construct prediction models with raw materials prices, it might provide the suggestions for the stakeholders that relate with the dry bulk market to make policy to avoid risk or reduce costs. The purpose of this study is not only using the price of raw materials prices to predict the dry bulk freight rate index but also BCI, BPI and BSI over the period January 2003 to December 2007 using X12, H-P filter, maximum R-square improvement (MAXR), nonlinear for predicting models with different freight rate (BDI, BCI, BPI and BSI). This research also introduces mean absolute percentage error (MAPE) values as a criterion to judge the model’s fitness in the different prediction models. The empirical results of those models show iron ore as the significant variable in this paper during the research period. Therefore, this paper adopts the iron ore to construct nonlinear models and predict dry bulk freight rate indexes. It also shows all the mean absolute percentage error (MAPE) values are below 10%. According to the MAPE value, it clearly shows these prediction models could be robust to predict the freight rate indexes and it also could provide useful information for stakeholders to avoid opera risk or deal their asset by portfolio.