Changes in Long Memory Feature and Informational Efficiency of Energy Commodity Markets

This study re-examines the degree of market efficiency in four major energy commodity markets (i.e., WTI, Brent, Gasoline, and Heating Oil). In particular, we explore the time-varying long memory features over time to assess the changes in informational efficiency of the markets. To investigate the degree of market efficiency, we considered the time-varying Hurst exponent by taking the rolling window approach with a time window of 1,008 observations. We found that the degree of long memory features and informational efficiency of four major energy commodity markets has changed over time. Although the estimated Hurst exponents exhibited a temporal movement towards market efficiency, energy commodity markets possessed a strong long memory property. The Hurst exponent showed an upward trend, implying that these markets are becoming more inefficient and predictable in recent years. In addition, both the daily and weekly returns generate the same pattern of time-varying Hurst exponent, referring to the fractal structure or self-similar structure in energy commodity markets.