Re: [Quantlib-users] question regarding using different day counters for rates and vol

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

Hello,
    you can read the thread at <
https://sourceforge.net/p/quantlib/mailman/quantlib-users/thread/CAKXkX9zpm1%2BKbrVthmNYe_qaEKjrmKo7WSixvvjVan0APnZfDw%40mail.gmail.com/#msg36015671>.
It refers to older engines, but the discussion also applies to
FdBlackScholesVanillaEngine.  It's a limitation of the engine,
unfortunately.  As a workaround, the suggestion was to create a new term
structure with the same day counter as the interest-rate curve and with
values of the volatility adjusted to account for the difference in day
count.  More details are in the thread, and a simple example is described
in <https://www.youtube.com/watch?v=BdM2Vs6R55g>.

Hope this helps,
    Luigi




On Fri, Mar 1, 2024 at 5:33 PM Kiki zzz <kik...@gm...> wrote:

> hi,
>
> I'm trying to use different day counters (e.g. Actual36525 and
> Business252) for interest rate and option volatility
> in FdBlackScholesVanillaEngine to price American or European option. Looks
> like the day counter in volatility does not affect option price. I
> think AnalyticEuropeanEngine on the other hand uses both day counters from
> interest rate and option volatility.
>
> My questions are
> - Is this the behavior by design (i.e. FdBlackScholesVanillaEngine only
> uses one day counter)?
> - If that's the case, what is the best way to workaround this? i.e. use
> different day counters for rates and vol in FdBlackScholesVanillaEngine.
>
> Thanks for the help in advance!
> Kiki
>
> _______________________________________________
> QuantLib-users mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>