How Long Memory Affects True Dependence Structure? Evidence From the Dow Jones Islamic Sub-indexes

Abstract

Long memory is a stylized fact found in most financial return series. In this paper, we seek to examine the impact of the presence of long memory on the dependence structure. First, we fit the multivariate dependence structure using R-vine copulas for pairs of raw and filtered returns. Second, we discuss whether or not changes in the fitted dependence may be driven by long memory. Based on the comparison results, we found that filtered returns of Dow Jones sub-indexes capture higher dependence in the conditional dependence structure. However, they capture lower dependence in bivariate dependence. Thus, the true degree of dependence is masked and such information is crucial for hedging and portfolios management.