DETERMINING THE DURATION OF CYCLES IN THE MARKET OF SECOND-HAND TANKER SHIPS, 1976–2001: IS PREDICTION POSSIBLE?

Abstract

In this paper we have calculated the duration of cycles manifested in the prices of second-hand tanker ships, using the method of Rescale Range (R/S) Analysis. We have considered a time series of 312 monthly data, taken between the years 1976–2001, and combined R/S with a study of its V-statistic, Power Spectrum Analysis and Sharpe Ratio. Our ultimate goal is to be able to help decision-makers use this analysis in order to buy or sell a ship at the "best timing". As the sale and purchase ship market exhibits a strongly deterministic cyclical behavior with a high degree of "irregularity" (or "volatility"), we expect the techniques of Nonlinear Dynamics and Chaos Theory to be appropriate to be applied to its dynamics. We have found evidence for the existence of two cycles of tanker ship prices, of duration 4 and 8 years respectively, in the years 1976–2001, in agreement with what is expected from the theoretical analysis of freight market cycles. We have shown that our time series is not a random walk process and have used software prediction methods to demonstrate that short term forecasting of second-hand tanker prices may indeed be possible. These results are, of course, preliminary since a lot more data is needed to test their forecasting validity and hence should only be regarded as a start in this interesting area of research.