Two robustness tests are performed to check the validity
of the above results. In the first test, the wavelet analysis
has been repeated by replacing the yields on 10-year
government bonds with a shorter-term interest rate such
as the yield on 1-year Treasury bills. For the vast majority
of industries the HTW-based correlation coefficients have
lower statistical significance, irrespective of the time horizon
considered, than those obtained from 10-year bond
rates. Moreover, the HTW cross-correlation estimated by
using 1 -year Treasury bill yields tends to be lower in magnitude
and statistical significance over the different time