Title: An empirical investigation of spot prices in tanker market using dynamic multiple regression models
Authors: Evangelos Xideas; Nikolaos D. Geomelos
Addresses: Department of Shipping Trade and Transport, Aegean University, Korai 2A, 82100 Chios, Greece. ' Department of Shipping Trade and Transport, Aegean University, Korai 2A, 82100 Chios, Greece
Abstract: The aim of this paper is to investigate spot prices of tanker market following an extensive empirical analysis using dynamic multiple regressions models via autoregressive distributed lags (ADL). These models constitute the most widely used pattern for empirical analysis since they are more suitable to ceteris paribus analysis offering much more flexibility in the formation of general functional form relationships. Also, they can be used to build better predicting models in order to support policy makers' decisions. This paper focuses in an extensive testing of stationarity and time trend issues in order to overcome the problem of spurious regression. The estimated models have been also used to generate ex-post and ex-ante forecasts in order to evaluate both econometric specification and predicting accuracy of dynamic multiple regression models. Results reveal that spot prices of each ship type market are affected by different sets of independent variables and the estimated models generate quite accurate predictions.
Keywords: empirical analysis; multiple regression models; autoregressive distributed lags; ADL; tanker spot prices; econometric modelling; stationarity; ex-post forecasting; ex-ante forecasting; tanker market; tankers.
International Journal of Decision Sciences, Risk and Management, 2011 Vol.3 No.3/4, pp.238 - 259
Received: 08 May 2021
Accepted: 12 May 2021
Published online: 22 Mar 2012 *