Long-Term Dependency Structure and Structural Breaks: Evidence from the U.S. Sector Returns and Volatility

Abstract

The study investigates the impact of structural breaks on the long memory of daily returns and variance of 11 sectors. Using multiple sequential structural breaks tests, we uncover numerous and roughly shared structural breaks. Results from two non-parametric, three semi-parametric, and three parametric fractional differencing models using break-adjusted and break-unadjusted returns reveal incidence of short memory and anti-persistence in sector returns. Regarding variance, we find that the removal of breaks from the sector series dampens the fractional differencing parameter estimates. Therefore, the observed long memory in variance may be attributable to the occurrence of structural breaks in the sector series.