Re: [Quantlib-users] Question about fitting SwaptionVolCube2 with SOFR swaptions

[Luigi B. <lui...@gm...>]

No, I don't think that would work.  VanillaSwap assumes a single fixing at
the beginning of the coupon.

On Fri, Feb 2, 2024 at 2:16 PM Martin Aussenhof <mau...@pe...>
wrote:

> Hi Luigi,
>
> Is it safe to use the VanillaSwap class to create a SOFR swaption? If so,
> is there an example somewhere. I’ve searched through the archive of this
> mailing list and found plenty of information, but am struggling to create a
> swaption based on sofr that correctly solves for normal vol.
>
> Thanks,
> Martin
>
> On Fri, 2 Feb 2024 at 12:25, Luigi Ballabio <lui...@gm...>
> wrote:
>
>> Hello Jason,
>>     SOFR-based swaptions are still an open implementation problem—see
>> https://github.com/lballabio/QuantLib/pull/1593 for an attempt at adding
>> them.  The vol cubes might need to be modified as well once we have them.
>> I don't think we have a corresponding swap index to pass right now.  I'm
>> not sure how this is managed in ORE; Peter, do you have any info on this?
>>
>> Luigi
>>
>>
>>
>> On Thu, Feb 1, 2024 at 7:55 PM Jason Lee <lee...@gm...> wrote:
>>
>>> Hi,
>>>
>>> For fitting SwaptionVolCube2, we need to specify the swapindexbase and
>>> shortswapindexbase.  While Libor is available, I used swapindex class, with
>>> familyname = usdliborswapisdafixa and tenor = 1y and 2y, as parameters.  My
>>> question is, most traded swaptions now are sofr1d based, how should I set
>>> up the related parameters in this case?  Also, some details on these two
>>> parameters would be highly appreciated.  I tried to research online but
>>> couldn't seem to find a good resource.  Is it related to the payment
>>> frequency on fix or float leg?
>>>
>>>
>>> Thank you so much!
>>>
>>> Best,
>>> Jason
>>> _______________________________________________
>>> QuantLib-users mailing list
>>> Qua...@li...
>>> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>>>
>> _______________________________________________
>> QuantLib-users mailing list
>> Qua...@li...
>> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>>
>