Re: [QuantLib-svn] SF.net SVN: quantlib:[16183] trunk/QuantLib

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: [QuantLib-svn] SF.net SVN: quantlib:[16183] trunk/QuantLib

Ferdinando Ametrano-4
Hi Chris

On Sun, Apr 19, 2009 at 5:19 PM,  <[hidden email]> wrote:
> Revision: 16183
>          http://quantlib.svn.sourceforge.net/quantlib/?rev=16183&view=rev
> Author:   chris_kenyon
> Date:     2009-04-19 15:19:25 +0000 (Sun, 19 Apr 2009)
> [...]
> Added Paths:
> -----------
>    trunk/QuantLib/ql/experimental/inflation/polynomial2D.hpp
>    trunk/QuantLib/ql/experimental/inflation/polynomial2Dspline.hpp

unless I've missed something you don't need Polynomial2DInterpolation
as you can obtain quadratic interpolation using the class
CubicInterpolation with DerivativeApprox::Parabolic.

Then it might be too exoteric, but it would benice to generalize
Interpolation2D to accept a policy (x_first or y_first) and (possibly)
different factories for the x and y axis, so that your new
Polynomial2DSpline (and the old BicubicSpline, BilinearInterpolation
too) would just become typedefs

ciao -- Nando

------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today.
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev
Reply | Threaded
Open this post in threaded view
|

Re: [QuantLib-svn] SF.net SVN: quantlib:[16183] trunk/QuantLib

Ferdinando Ametrano-4
Hi Chris

following up to my previous email (quoted below) where I suggested you
don't need Polynomial2DInterpolation I've just committed this change
in Rev16320. Please review it.

btw Since your Rev16183 commit the testsuite has linking warnings that
should be fixed:

c:\projects\trunk\quantlib\ql\experimental\inflation\interpolatedyoyoptionletstripper.hpp(108)
: warning C4244: 'argument' : conversion from 'QuantLib::Rate' to
'QuantLib::Size', possible loss of data
        c:\projects\trunk\quantlib\ql\experimental\inflation\interpolatedyoyoptionletstripper.hpp(106)
: while compiling class template member function
'QuantLib::InterpolatedYoYOptionletStripper<Interpolator1D>::ObjectiveFunction::ObjectiveFunction(QuantLib::YoYInflationCapFloor::Type,QuantLib::Real,QuantLib::Rate,QuantLib::Period
&,QuantLib::Natural,boost::shared_ptr<T>,const
boost::shared_ptr<QuantLib::YoYCapFloorTermPriceSurface> &,const
boost::shared_ptr<QuantLib::YoYInflationCapFloorEngine>
&,QuantLib::Real)'
        with
        [
            Interpolator1D=QuantLib::Linear,
            T=QuantLib::YoYInflationIndex
        ]
        c:\projects\trunk\quantlib\ql\experimental\inflation\interpolatedyoyoptionletstripper.hpp(213)
: see reference to class template instantiation
'QuantLib::InterpolatedYoYOptionletStripper<Interpolator1D>::ObjectiveFunction'
being compiled
        with
        [
            Interpolator1D=QuantLib::Linear
        ]
        c:\projects\trunk\quantlib\ql\experimental\inflation\interpolatedyoyoptionletstripper.hpp(167)
: while compiling class template member function 'void
QuantLib::InterpolatedYoYOptionletStripper<Interpolator1D>::initialize(const
boost::shared_ptr<T> &,const
boost::shared_ptr<QuantLib::YoYInflationCapFloorEngine> &,const
QuantLib::Real) const'
        with
        [
            Interpolator1D=QuantLib::Linear,
            T=QuantLib::YoYCapFloorTermPriceSurface
        ]
        c:\projects\trunk\quantlib\test-suite\inflationvol.cpp(245) :
see reference to class template instantiation
'QuantLib::InterpolatedYoYOptionletStripper<Interpolator1D>' being
compiled
        with
        [
            Interpolator1D=QuantLib::Linear
        ]
c:\projects\trunk\quantlib\ql\experimental\inflation\interpolatedyoyoptionletstripper.hpp(117)
: warning C4244: 'argument' : conversion from 'QuantLib::Rate' to
'QuantLib::Size', possible loss of data
c:\projects\trunk\quantlib\ql\experimental\inflation\interpolatedyoyoptionletstripper.hpp(123)
: warning C4244: 'argument' : conversion from 'QuantLib::Time' to
'QuantLib::Integer', possible loss of data
c:\projects\trunk\quantlib\ql\experimental\inflation\interpolatedyoyoptionletstripper.hpp(131)
: warning C4244: 'initializing' : conversion from 'double' to
'QuantLib::Size', possible loss of data


ciao -- Nando

On Mon, Apr 20, 2009 at 3:26 PM, Ferdinando Ametrano<[hidden email]> wrote:

> Hi Chris
>
> On Sun, Apr 19, 2009 at 5:19 PM,  <[hidden email]> wrote:
>> Revision: 16183
>>          http://quantlib.svn.sourceforge.net/quantlib/?rev=16183&view=rev
>> Author:   chris_kenyon
>> Date:     2009-04-19 15:19:25 +0000 (Sun, 19 Apr 2009)
>> [...]
>> Added Paths:
>> -----------
>>    trunk/QuantLib/ql/experimental/inflation/polynomial2D.hpp
>>    trunk/QuantLib/ql/experimental/inflation/polynomial2Dspline.hpp
>
> unless I've missed something you don't need Polynomial2DInterpolation
> as you can obtain quadratic interpolation using the class
> CubicInterpolation with DerivativeApprox::Parabolic.
>
> Then it might be too exoteric, but it would benice to generalize
> Interpolation2D to accept a policy (x_first or y_first) and (possibly)
> different factories for the x and y axis, so that your new
> Polynomial2DSpline (and the old BicubicSpline, BilinearInterpolation
> too) would just become typedefs
>
> ciao -- Nando
>

------------------------------------------------------------------------------
_______________________________________________
QuantLib-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-dev