Predictive Accuracy of Crude Oil Futures Prices

Abstract

Oil is at the moment one of the most important energy resources in the world. Therefore price of oil and oil related products are observed by many firms and individuals every day. Oil price changes have influence on capital budgeting plans and foreign value asset investments. Sales and profits of many major industries are depended from the price fluctuation of crude oil.

The purpose of this study is to evaluate predicting power of crude oil futures prices in short term i.e. a few days and to test if the US & UK futures markets are efficient in the way that the Efficient Market Hypothesis states. Possible predicting power of crude oil futures prices is analyzed to see if it could be possible to predict the price of crude oil in the near future. The data covers period from 15th of April 2004 to November 15th 2006 and two crude oil futures which are tested against Brent crude oil price from Intercontinental Exchange in London. The empirical test is done using Granger causality test in order to analyze the potential lead-lag relationships of the time series.

Two hypotheses were formed in this study based on the findings in earlier research. The first hypothesis is that the futures prices are able to forecast the future price of crude oil in near term. The second hypothesis is that the Efficient Market Hypothesis holds between The US and UK market. Empirical tests confirmed that crude oil futures prices lead the price of crude oil in short time periods; meaning roughly one to four days. After four days the effect fades away. More accurate models should be developed in order to get more accurate results for price prediction.