Long memory in energy futures prices

This paper extends the work in Serletis [Serletis, A. (1992). Unit root behavior in energy futures prices. The Energy Journal 13, 119–128] by re-examining the empirical evidence for random walk type behavior in energy futures prices. It tests for fractional integrating dynamics in energy futures markets utilizing more recent data (from January 3, 1994 to June 30, 2005) and a new semi-parametric wavelet-based estimator, which is superior to the more prevalent GPH estimator (on the basis of Monte-Carlo evidence). We find new evidence that energy prices display long memory and that the particular form of long memory is anti-persistence, characterized by the variance of each series being dominated by high frequency (low wavelet scale) components.