Premium
A statistical study of the short‐ and long‐term drivers of crude oil prices
Author(s) -
Hassani Hossein,
Moudassir Mouhamad,
Christodoulides Pantelis,
Odulaja Adedapo
Publication year - 2017
Publication title -
opec energy review
Language(s) - English
Resource type - Journals
eISSN - 1753-0237
pISSN - 1753-0229
DOI - 10.1111/opec.12095
Subject(s) - econometrics , crude oil , term (time) , error correction model , vector autoregression , economics , autoregressive model , futures contract , multivariate statistics , cointegration , financial economics , statistics , mathematics , engineering , physics , quantum mechanics , petroleum engineering
The paper statistically examines whether the fluctuations in crude oil prices over the last 10 years are due to speculative activities or changes in market fundamentals. Following a brief review of the oil market developments and speculative activities in recent years, the paper employs two multivariate time series models—the vector autoregressive model (VAR) and the vector error correction model (VECM)—to evaluate observed fluctuations in crude oil prices. The VAR and its granger test allow for the detection of factors responsible for fluctuations in crude oil prices, while the VECM tests the co‐integration of the variables. The VAR includes both fundamental and non‐fundamental variables as covariates. The VECM results show that in both the short‐ and long‐term the significant variables of the VAR model are co‐integrated. The empirical results confirm that in the long‐term the fundamentals are the main drivers of crude oil prices. However, in the short term, both fundamentals and speculative positions in the oil futures market differentially cause fluctuations in crude oil prices. The effect of the fundamentals is more than double in the long term, when compared to the short‐term.