Posted by
Peter Caspers-4 on
Nov 15, 2015; 5:19pm
URL: http://quantlib.414.s1.nabble.com/Possible-Swaption-Code-Bug-tp17075p17085.html
Hi Etuka,
an easy test would be to set up a Bermudan swaption with one exercise
(say after 2 years), the underlying being in the money (e.g. pay fixed
2%. receive Euribor 6M, use a flat yield term structure with forward
at 3% for forwarding and discounting) and setting the model volatility
to zero. Then compare the swaption NPV from the tree engine with the
exercise-into swap NPV (swap with 2y forward start into 3y) and the
whole swap NPV (spot into 5y), both produced with a
DiscountingSwapEngine, on the same curve. Let us know the results.
Best regards
Peter
On 13 November 2015 at 17:41, Etuka Onono
<
[hidden email]> wrote:
> Hello.
>
> I think there is a bug in the implementation of Bermudan Swaption pricers that depend on the class DiscretizedSwaption. If you have a 5 year swap, resetting in each of its 5 years, then the bermudan swaption should allow you to decide whether to enter a 5, 4, 3, 2 or 1 year swap terminating in the 5th year. It looks to me more like, in the current implementation, a holder accrues the value of the whole swap in the tree, whenever the swaption is exercised.
>
> Grateful for feedback or confirmation. Let me know if there is anything I can do to help follow up.
>
> Regards,
> Etuka
> ------------------------------------------------------------------------------
> _______________________________________________
> QuantLib-users mailing list
>
[hidden email]
>
https://lists.sourceforge.net/lists/listinfo/quantlib-users------------------------------------------------------------------------------
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users