Re: [Quantlib-users] Option does not decay in time

[Dan K. <dan...@gm...>]

Thanks! I very quickly tried to add a theta calculation to
pricingengines/vanilla/fdstepconditionengine.hpp, but it gave crazy results.
Here is the non-working code I added at the end of
calculate(PricingEngine::results*):

Time dt=0.001;
model.rollback(arraySet, getResidualTime(),-dt, timeSteps_, conditionSet);

SampledCurve earlyPrices(gridPoints_);
SampledCurve earlyControl(gridPoints_);
earlyPrices.values() = arraySet[0];
earlyControl.values() = arraySet[1];

Real earlyValue=earlyPrices.valueAtCenter()- earlyControl.valueAtCenter()+
black.value();

results->theta = (results->value - earlyValue)/dt;

Am I not evolving the prices back in time correctly? I leave this here in
case anybody wants to fix this and commit since it could be generally
useful.

Dan

Dan

On Wed, Mar 23, 2011 at 2:52 AM, Luigi Ballabio <lui...@gm...>wrote:

>
> Dan,
>        you probably gave a fixed reference date to your curves.  Use the
> constructors taking a calendar and a number of settlement days instead.
>
> Luigi
>
>
> On Tue, 2011-03-22 at 23:42 -0500, Dan Krop wrote:
> > Hello Quantlibbers,
> >    I am having a problem approximating theta because it seems the
> > option I have has constant NPV over time for a given underlying
> > price.
> >
> >
> > Here is a test function,
> >
> >
> > Real approximateTheta(VanillaOption & option)
> > {
> > DayCounter dc = Actual365Fixed();
> >
> >
> > Date today = Settings::instance().evaluationDate();
> > Time dT = dc.yearFraction(today-5,today+5);
> >
> >
> > std::cout<<"today="<<today<<"  dT="<<dT<<std::endl;
> >
> > Settings::instance().evaluationDate()=today-5;
> > Real value_m=option.NPV();
> >
> > Settings::instance().evaluationDate()=today+5;
> > Real value_p=option.NPV();
> >
> >
> > Settings::instance().evaluationDate()=today;
> > std::cout<<"m="<<value_m<<" p="<<value_p<<std::endl;
> >
> >
> > Real theta=(value_p - value_m)/dT;
> > return theta;
> > }
> >
> >
> > When I run this, I get that value_p=value_m. The option uses the
> > AnalyticEuropeanEngine. I thought that options registered with the
> > evaluationDate, so simply moving the date should be enough to get them
> > to decay. I must be doing something wrong. I can provide a more
> > complete example if it will help.
> >
> >
> > Thanks in advance,
> >      Dan
> >
> ------------------------------------------------------------------------------
> > Enable your software for Intel(R) Active Management Technology to meet
> the
> > growing manageability and security demands of your customers. Businesses
> > are taking advantage of Intel(R) vPro (TM) technology - will your
> software
> > be a part of the solution? Download the Intel(R) Manageability Checker
> > today! http://p.sf.net/sfu/intel-dev2devmar
> > _______________________________________________ QuantLib-users mailing
> list Qua...@li...
> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>
> --
>
> There are two ways of constructing a software design. One way is to
> make it so simple that there are obviously no deficiencies. And the
> other way is to make it so complicated that there are no obvious
> deficiencies.
> -- C. A. R. Hoare
>
>
>