Re: [Quantlib-users] What are some examples of moving yield term structures?

[Wei Li <ttl...@gm...>]

Hello Luigi,

Thank you for your reply. I guess I didn't look through all the
constructors so I missed the ones with settlementDays as input. I have
another question though: is there any workaround to make the
InterpolatedDiscountCurve reference-date-dependent (like I mentioned, we
construct some curves by giving discount factors and the discounting dates
that we calculated somewhere else)? I remember I have read from the
maillist that this is not supported. But I also see that there are some
constructors with settlementDays as input if the constructors of this
class, although they are marked as protected and they do not take discount
factors as input. If that is not possible, is there some rate helper class
that can construct zero coupon bonds from the discount factors and use
those instruments to construct the curve?

Cheers,
Wei

On Mon, Nov 4, 2024 at 6:40 PM Luigi Ballabio <lui...@gm...>
wrote:

> Hello Wei Li,
>     PiecewiseYieldCurve can be moving or not, depending on how you build
> it.  If you're passing a reference date as the first argument to its
> constructor, then the curve will stay fixed at that reference date.  If
> you're passing a number of days and a calendar instead, the curve will move
> with the evaluation date.
>
> Luigi
>
>
> On Fri, Nov 1, 2024 at 11:20 AM Wei Li <ttl...@gm...> wrote:
>
>> Dear all,
>>
>> When I am developing using ql, I learned that with *moving *term
>> structures (whose moving_ property are set to true and are basically
>> constructed without using specific dates), I can calculate the theta per
>> day measurement by bumping the evaluation date to the next business day.
>>
>> However, I am not sure what are the examples of such moving term
>> structures. So far our project uses exclusively *InterpolatedDiscountCurve
>> *and *PiecewiseYieldCurve : *we construct the first class by giving the
>> specific dates and discount factors (calculated somewhere else) on such
>> dates, and construct the second class with market quotes of different types
>> of instruments and do the bootstrapping on the fly. As far as I know,
>> neither of these two is a moving term structure. As a result, I can't get
>> the expected theta per day by just bumping the evaluation date. Our
>> portfolio is mainly composed of FX and fixed income exotic options and the
>> hedging instruments so the valuation relies heavily on the yield term
>> structures.
>>
>> So I was wondering, if we want to get such theta per day measurement,
>> what would be the proper classes that we should use? Thank you very much!
>>
>> Cheers,
>> Wei
>>
>>
>> _______________________________________________
>> QuantLib-users mailing list
>> Qua...@li...
>> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>>
>