Analysis of shipping cycles and their predictability
View/
Keywords
Shipping cycles ; Maritime economics ; Forecasting models ; Econometric analysis ; Machine learning ; Freight rates ; Shipbuilding ; Scrapping cycles ; Technological innovation ; Environmental regulation ; Decarbonization ; Market predictability ; Maritime finance ; Global tradeAbstract
It examines the pattern predictability and forecast ability of shipping cycles over a period from the 19th to the 21st century, noting them in order to demonstrate that these become more complex as the global economy accelerates. Based on a comprehensive review of the literature, this study identifies the key drivers (global GDP growth, trade flows wind and shipbuilding activity) of maritime cycles alongside an evolving set of cyclical amplifier variables (financial macro factors) as well as new causative factors that add to driving fluctuations in shipping markets: technological change and environmental policy intervention. The results show that shipping cycles still are fundamentally cyclical but more and more multi‐dimensional as they are influenced not only by traditional economic forces, but also by contemporary digitalization, automation and decarbonization trends.
The analysis reveals that econometric and time series models (e.g., ARIMA, VAR, GARCH) provide analytical short-term predictive accuracy but lose effectiveness during structural turning points or crises. On the other hand, machine learning and hybrid models are able to improve predictability capturing nonlinear relationships as well as adjusting to market volatilities. However, challenges of data sharing, model interpretability and real-time information integration are still imminent.
The research also reveals key knowledge gaps within the literature, such as few behavioral models being available, lack of use of big data and lack of emphasis on sustainability-led cycles. Instead, it suggests a more holistic, interdisciplinary approach to forecasting with economic, behavioral and environmental variables. The findings highlight that full predictability of shipping cycles continues to elude us, but the use of AI, data analytics and policy coordination offer near-term opportunities to greatly enhance resilience and foresight in global maritime markets.
Postgraduate Studies Programme
Shipping ManagementDepartment
Σχολή Ναυτιλίας και Βιομηχανίας. Τμήμα Ναυτιλιακών ΣπουδώνNumber of pages
83Language
EnglishCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Όχι Παράγωγα Έργα 3.0 Ελλάδα