Re: [Quantlib-dev] IborIndex::forecastFixing

[Francois B. <ig...@gm...>]

Hi Peter,

Maybe some context. I was looking at the FraRateHelper class and I see that
it uses an instance of IborIndex. Is the assumption of a linear compound
rate valid for the FraRateHelper classes too?

My ultimate goal is to create my own subclass of RateHelper that accepts a
quote of a forward rate, start date and end date so that I can build a
PiecewiseYieldCurve and pass this new ratehelper subclass. See my other
post about this. If that assumption about the IborIndex is always true,
then I won't be able to use it.

thanks
Francois Botha


On Mon, 30 Jul 2018 at 21:34, Peter Caspers <pca...@gm...> wrote:

> Hi Francois,
>
> let’s be a bit careful changing this ;-) … this computes a projected
> fixing of an Ibor Index which by definition is a simply (linearly)
> compounded rate r. The forward compound factor between d1 and d2 is
> therefore
>
> 1 + r * yearFraction(d1, d2)
>
> The same compound factor in terms of discount factors (and therefore
> independent of any rate conventions) is given by disc1 / disc2, from which
> the return value formula follows.
>
> Best Regards
> Peter
>
>
>
> On 30 Jul 2018, at 13:02, Francois Botha <ig...@gm...> wrote:
>
> Hi,
>
> I notice this code:
>
>     inline Rate IborIndex::forecastFixing(const Date& d1,
>                                           const Date& d2,
>                                           Time t) const {
>         QL_REQUIRE(!termStructure_.empty(),
>                    "null term structure set to this instance of " <<
> name());
>         DiscountFactor disc1 = termStructure_->discount(d1);
>         DiscountFactor disc2 = termStructure_->discount(d2);
>         return (disc1/disc2 - 1.0) / t;
>     }
>
> The last line looks odd to me. disc1/disc would return a non-continuously
> compounded return. Dividing by t is the approach to annualise continuous
> returns. So the 2 are not consistent.
>
> I would have expected:
>     return std::pow(disc1 / disc2, 1 / t) - 1.0;
>
> On the other hand, if I make that change, many tests fail, so I'm by no
> means confident about my assertion above.
>
> regards
> Francois Botha
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org <http://slashdot.org>!
> http://sdm.link/slashdot_______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>
>
>