Search

The long memory behavior of gold returns during the post-Bretton Woods period is analyzed through a new rescaled range method. Weekly spot prices in the London gold market are subjected to the new rescaled range analysis. Statistical findings indicate significant instability in the long memory behavior of gold returns. In particular, the gold price series during the years before or after and excluding 1979 showed no substantial evidence of long memory. A closer investigation revealed that once some observations relative to major political happenings in the Middle East, together with the Hunts event, in the latter part of 1979 are omitted, there is no more substantial evidence of long memory behavior in gold returns. This implies that uncommon happenings during that year may be credited for the false findings of long memory behavior in gold returns.