Quantitative analysis of the dry bulk freight market, including forecasting and decision making
KeywordsFreight market fundamentals ; Forecasting model ; Spot freight market ; Decision support systems ; Trip time charter ; Dry Bulk Economic Climate Index (DBECI) ; Ναυλώσεις
This thesis provides a quantitative analysis of the dry bulk freight market and investigates a number of crucial issues associated with ship chartering. To this end, it analyzes the dynamics of spot and period freight markets separately and leads to insightful conclusions. At the outset, a theoretical discussion of the market fundamentals sets the scene and is followed by a thorough exploration of the key determinants of freight rates. At this point a new composite indicator is constructed -the Dry Bulk Economic Climate Index (DBECI)-, which is tailored to the dry cargo market and mirrors the aggregate impact of some carefully selected economic variables. As opposed to conventional approaches, the structure and the weighting scheme of this index are based on extensive exploratory and numerical analysis. This enhances the credibility of the DBECI and ultimately gives rise to more meaningful analyses. The next step is to carry out Co-integration analysis, Granger Causality tests and Impulse Response analysis in order to identify possible linkages between this new indicator and the freight market. In parallel, a similar numerical analysis is performed for some equally important determinants, such as the Chinese steel production, the average bunker prices, the port congestion, and the price of the most traded bulk commodities. The results reveal significant lead-lag relationships for the cases of DBECI, bunker prices, Chinese steel production, and commodity prices, while the port congestion appears to lead the freight rates only in the Capesize sector. The subsequent section is devoted to the development of parsimonious multivariate forecasting models (VAR/VECM and VARX). In this respect, the preceding theoretical and empirical analysis constitutes the groundwork for the selection of the most appropriate explanatory variables. Specifically, the Chinese steel production and the fleet development are used as endogenous variables, while the DBECI and the fuel prices are treated as exogenous. Next, a univariate framework (ARIMA) for the freight rates of Panamax and Capesize vessels is developed and serves as a benchmark for comparison of the forecasting accuracy of the proposed multivariate models. The findings show that the VARX model outperforms both of the alternative approaches, suggesting that the incorporation of these two exogenous variables (DBECI and average bunker prices) can significantly enhance the robustness of simpler models and ultimately result in more accurate forecasts. The last part of this thesis involves the investigation of excess return opportunities in the spot market. For this purpose the dynamics between trip charters and their corresponding voyage charters are studied. The thesis first examines the existence of a long-run equilibrium relationship and then develops a new methodology based on technical analysis so as to identify excess return signals and formulate a suitable chartering strategy. The results reveal that this approach outperforms the ‘naïve’ strategy of always chartering in vessels on trip time charters and perform the underlying voyage charters. Overall, the present thesis is of interest to academics and maritime practitioners alike. It fills significant gaps in the literature, while at the same time it can serve as a powerful decision support tool for shipping companies.