Periodical characteristics of shipbuilding market activity: A wavelet analysis

Li Song, Jihong Chen, Kevin X. Li, Xiang Liu, Yijie Fei, Hang Yu

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Shipbuilding, a significant sector of the international shipping market, shows fluctuations and periodicity typical of variations in supply and demand. However, it takes a significant amount of time to build a ship, and the influencing factors are complex. Therefore, attempts to understand the market's fluctuations by directly analyzing these influencing factors suffer from high inaccuracy, necessitating a quantitative analysis of shipbuilding's periodic features. In this paper, we used wavelet analysis, an efficient way to analyze time series data, to analyze the unit price of a Panamax bulk ship's compensated gross tonnage and get the periodical features of the market. Choosing a significantly periodic wavelet coefficient curve yields three different cycle lengths: a 1-year seasonal cycle, a 3.5-year short-term cycle, and a 13-year medium-to-long-term cycle. Finally, we analyzed the accumulated wavelet coefficient curve, and forecasted that the market should reach its next prosperity phase around 2023 in the medium-to-long-term cycle. The present study is in the direct interest of maritime practitioners, because it helps to more precisely forecast shipbuilding market fluctuations, allowing them to make informed decisions.

Original languageEnglish (US)
Pages (from-to)692-702
Number of pages11
JournalJournal of Marine Science and Technology (Taiwan)
Volume26
Issue number5
DOIs
StatePublished - 2018

All Science Journal Classification (ASJC) codes

  • Oceanography
  • Ocean Engineering
  • Mechanics of Materials
  • Mechanical Engineering

Keywords

  • Cycle
  • Forecast
  • Shipbuilding market
  • Shipping
  • Wavelet analysis

Fingerprint

Dive into the research topics of 'Periodical characteristics of shipbuilding market activity: A wavelet analysis'. Together they form a unique fingerprint.

Cite this