The efficiency of the oil futures markets
: information, price discovery and long memory

  • Saada Abdullahi

Student thesis: Doctoral Thesis


This thesis investigates the efficiency of the crude oil futures markets by addressing four important issues using different theoretical and methodological perspectives. Data of different frequencies was employed in the analysis covering the period 2000 to 2011. First, the short and long term efficiency is examined by testing the unbiasedness of the oil futures price in predicting the expected spot price using the Johansen (1988) and the Engle-Granger (1987) cointegration tests, and the Error Correction Model (ECM). The results suggest that the oil futures markets are unbiased in the long term but not in the short term, and the inefficiency is not caused by the time-varying risk premium. The results also show that the oil futures market are unbiased in the multi-contract and multi-market framework but not in all maturities. Second, the price discovery relationship between the oil spot and futures markets and across contract is investigated by employing the Vector Error Correction Model (VECM), Gonzalo-Granger (1995) common factor weight approach and the Garbade-Silber (1983) short run dynamic model. Empirical results indicate that price discovery is initiated in the futures market because it impounds more information than the spot market. However, the results of the cross-contract analysis show that the three-month futures contract leads one-month contract in price discovery while the relationship changes in the short term. Third, the price change and trading volume relationship is examined in the oil futures markets using the generalized method of moments (GMM), Granger causality test, impulse response function and variance decomposition approaches. The findings reject the postulation of a positive relationship between price change and trading volume, suggesting that they are not driven by the same information. Additionally, the results suggest that trading volume cannot predict price changes in all the oil markets. Lastly, this thesis investigates long memory in the oil futures return using the GARCH models, and the results indicate that both the short and long memory models support predictability in returns which violates the weak form efficient hypothesis. In sum, the findings provide new evidence on the informational efficiency of the international oil futures markets, which have significant implications for hedgers, speculators, financial analysts and policymakers. The thesis recommends that market participants and regulators should look at various aspects of these markets for effective strategies and policy implementation.
Date of AwardSep 2012
Original languageEnglish
SponsorsPetroleum Technology Development Fund (PTDF)
SupervisorReza Kouhy (Supervisor), Heather Tarbert (Supervisor) & Zahid Muhammad (Supervisor)

Cite this