Re: [Quantlib-dev] How to construct an interest curve using an existing curve with two minor modifications?

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

I'm not sure that the video applies, unfortunately.  It looks like the
inverse case of https://youtu.be/8Lc5r0YxAME : given a curve with reference
on today's date, imply an earlier curve.  The 0.98 discount factor would be
the factor from today's date to a couple of days earlier, which can't be
implied from the current curve.

I think the way to do it would be similar to what we did in the
ImpliedTermStructure class.  You'll need to write a new class (you can copy
ImpliedTermStructure and modify it) that takes the original curve, the new
reference date, and the discount factor between the new reference and
today.  Its discountImpl method would figure out the time according to the
original curve and return the original discount multiplied by 0.98 (or
whatever the discount factor is between the two reference dates).

Hope this helps,
    Luigi


On Thu, Jan 11, 2018 at 8:44 PM Francois Botha <ig...@gm...> wrote:

> Hi
>
> I hope Luigi agrees with me. For point 1, I think you should look at a
> video by the master himself. https://youtu.be/pc1yOmxU2GQ
>
> For point 2, I'm not sure whether it's possible or why you would want to
> do that. Scaling the discount rate by a fixed factor would imply for
> instance that the new curve would have a discount rate of 0.98 at time 0.
> To me, that doesn't make sense. A discount rate should always be 1 at time
> 0. But maybe I'm wrong.
>
> Regards
> Francois
>
>
> On 11 Jan 2018 15:04, "zcg6433" <yuz...@gm...> wrote:
>
> How to construct an zero curve using an existing curve with the following
> two
> modifications?
> 1. move the reference date by two dates earlier.
> 2. Assuming the knot points of the curve are represented using discount
> factors, then the all of the knot points of the new curve should be 0.98 *
> those of the old curve. ( i.e. this has the equivalent effect of
> downscaling
> the discount factor of the original curve by 0.98 )
>
>
>
> --
> Sent from: http://quantlib.10058.n7.nabble.com/quantlib-dev-f8818.html
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> QuantLib-dev mailing list
> Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-dev
>