An econometric model of the one million barrel tanker market

Omosola, Afolabi Akin (1999) An econometric model of the one million barrel tanker market. Doctoral thesis, London Guildhall University.

[img]
Preview
Text
312932.pdf - Published Version

Download (18MB) | Preview

Abstract / Description

This thesis presents a disaggregated econometric model and set of forecasts of the supply of one million barrel tankers (l00-160,000DWT). The model examined the factors of tanker demand and supply. It looked at the oil imports Western Europe and USA as sources of demand for tankers in this category. It examined tanker demand, freight rate index, time charter index, new order, newbuilding prices, deliveries and scrapping.

The reasons behind this research were enumerated. This is to forecast supply in order to avoid the problem that caused depression of tanker market in the past. The surplus tonnage that made the tanker freight rate collapse in the 70's. Further more, it examined and speculates on the effect of the USA Oil Pollution Act 1990. The background problem of the oil tanker market was examined; oil and tanker markets were analysed. The thesis looks at the effect of oil imports and exports on oil tankers. It examined the structure of tanker market and looked at the product and the market players. The routes and employment of these tankers were examined; these are crucial to tanker supply.

The thesis examined forecasting methods and techniques. It considered past attempts at modelling the shipping markets. It is argued that the structure of existing models of the shipping market is theoretically flawed and also that the existing model is aggregated. These models are freight rate models rather than tanker supply. The forecasting of one million barrel tankers has never been implemented before. Another argument is that any of these forecasting techniques can be used in forecasting oil tanker supply. In addition because the previous models did not test for unit roots in variables and the long run equilibrium of the series, the thesis has tested for these. The time series data are in most case non-stationary and should be made stationary before estimation.

The model of tanker supply was presented to show the behaviour of the market. The assumption is that oil import in Western Europe and USA will be crucial to one million barrel oil tankers. The relationship of factors that affects tanker supply was enumerated and analysed. The tanker supply is an identity equation where deliveries and scrapping play a crucial role. The deliveries will cause increase to supply while scrapping will cause it to reduce. The simultaneous equation model was identified using order and rank condition of identification. The unit roots and cointegration of the series were computed and analysed. The series were made stationary and first differences of the series were used in the model. The series are also cointegrated, showing a long run relationship.

The econometric version of the theoretical model was estimated. The model is then used in simulating the dynamic response of tanker market to anticipated and unanticipated external shocks. The market plays a crucial role in the adjustment process. The estimated model was used to forecast the supply of one million barrel tankers. Ten year forecasts of the dependent factors of the model were produced under a number of different assumptions.

The thesis examined and compared the results of the alternative forecasting models with the econometric one. The univariate autoregressive integrated moving average, random walk and exponential smoothing were estimated and analysed. The accuracy of the forecast was tested; using E, RSE, RPE and Theil, with emphasis on RSE.

Item Type: Thesis (Doctoral)
Additional Information: uk.bl.ethos.312932
Uncontrolled Keywords: ships; tankers; million barrel capacity; supply and demand; disaggregated econometric models; forecasts
Subjects: 300 Social sciences > 380 Commerce, communications & transportation
Department: Guildhall School of Business and Law
Depositing User: Mary Burslem
Date Deposited: 28 Mar 2022 12:53
Last Modified: 28 Mar 2022 12:53
URI: http://repository.londonmet.ac.uk/id/eprint/7312

Downloads

Downloads per month over past year



Downloads each year

Actions (login required)

View Item View Item