Re: Fwd: Error dereferencing handle (for fixings in an index)

Posted by Luigi Ballabio on Jun 15, 2012; 2:16pm
URL: http://quantlib.414.s1.nabble.com/Fwd-Error-dereferencing-handle-for-fixings-in-an-index-tp8274p8277.html

Adding fixings only works for fixings in the past; the index will
always try to forecast future fixings on a curve, ignoring fixings
added for future dates.

If you look at Examples/Bonds/Bonds.cpp in the C++ library, you'll
find an example of building a Libor forecast curve bootstrapped on a
set of deposit and swap rates.  The example is from the old days when
there was just one Libor curve; nowadays you'll have to be careful
what deposits, futures or swaps to include depending on the Libor (for
instance, see Ferdinando's slides on the subject available at
<http://www.statpro.com/quantlib_forum/quantlib_presentations.aspx>).

Luigi

On Wed, Jun 13, 2012 at 1:22 PM, rohan talwar <[hidden email]> wrote:

> Thanks for the swift response - I don't but file at work - I pretty much
> copy the implementation of the FRN example .
>
> Perhaps I am constructing the index (Ibor) incorrectly. If this is most
> likely issue I will go back and review the model. If I still have issue I
> will send worksheet (file at work presently).
>
> What are the suitable term structures one can feed in, given we need to
> extract forward libor settings for the frn with fixed spread, and perhaps
> where the market spread has moved (i.e. NPV not = par).
>
> Are there known issues with sending in a time series into the
> "=qlIndexAddFixings2()" xl call?
>
> Cheers, Rohan
>
>
>
> ________________________________
> From: Luigi Ballabio <[hidden email]>
> To: Rohan TALWAR <[hidden email]>
> Cc: quantlib-users <[hidden email]>
> Sent: Wednesday, 13 June 2012, 20:13
> Subject: Re: [Quantlib-users] Fwd: Error dereferencing handle (for fixings
> in an index)
>
> If your error is that a null handle is dereferenced, then past fixings
> are not involved.
> You're probably missing the term structure needed to forecast future
> fixings.  Do you pass a yield term-structure handle to the index when
> you build it?  Do you link it to an actual term structure before
> calculating the NPV?
>
> Luigi
>
>
> On Wed, Jun 13, 2012 at 11:49 AM, Rohan TALWAR <[hidden email]>
> wrote:
>> Hi
>>
>> I have tried to trace back the error dereferencing the handle problem but
>> it
>> seems to relate to code examples not QL XL.
>> I am trying to put a worksheet together to calc an FRN and the fixings are
>> added to an index as a time series. However, when trying to get the
>> accruals
>> and also NPV which needs the set leg it seems, it falls over. I can send
>> the
>> worksheet if required (it's a bit messy).
>> I've looked at the examples, but these don't relate to where you have a
>> series of fixings, only if you ad an individual fixing.
>>
>> Any ideas would be greatly appreciated.
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Live Security Virtual Conference
>> Exclusive live event will cover all the ways today's security and
>> threat landscape has changed and how IT managers can respond. Discussions
>> will include endpoint security, mobile security and the latest in malware
>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> _______________________________________________
>> QuantLib-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>>
>
>

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users