Hi,
I use the SwaptionVolCube1 class as a basis for an own cube, essentially
replacing the sabr smile section by another one. I noticed when calibrating
the free parameters of the cube (corresponding to beta in case of the
original class) to the cms market using the recalibration() method, these
parameters remain unchanged for swap tenors not involved in calibration
(because not quoted). Although not explicitly tested I assume that this is
also true for the original cube. E.g. (in terms of SwaptionVolCube1) when
the betas for the 2y and 5y underlying are recalibrated, the betas for 3y
and 4y are not updated (linearly interpolated between 2y and 5y). (At least
when using isAtmCalibrated=true, did not try the sparse parameter approach).
Now my questions are:
1. Is the observation correct?
2. Is the behaviour meant like that?
I have a done a modification to my cube in order to have the (from my point
of view) correct behaviour, so if 1=yes, 2=no, I would be happy to translate
it to SwaptionVolCube1 and provide it.
Thanks and regards
Peter
------------------------------------------------------------------------------
vRanger cuts backup time in half-while increasing security.
With the market-leading solution for virtual backup and recovery,
you get blazing-fast, flexible, and affordable data protection.
Download your free trial now.
http://p.sf.net/sfu/quest-d2dcopy1_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev
Locked
