A same company has two different bonds. I expected the Z-spread to be close for both bonds (since my representation of the Z-spread is the spread due to credit-risk proper to the company).
Here is an example:
Altice France SA/France => 95.282 Z-spread for a bond.
Altice France SA/France => 473.470 Z-spread for an other bond.
Looking at a whole set of data, I observe that it is not rare to have significative different spreads.
Should I conclude that the Z-spread model has a bond-dependancy ?
Shouldn't I conclude that a better model could be a time-dependant spread (I could build it via bootstraping using increasing maturities of a set of bonds of a same company ?).
Thanks a lot for your feedbacks !