Re: trouble with the finite differences framework

Posted by Luigi Ballabio-2 on
URL: http://quantlib.414.s1.nabble.com/trouble-with-the-finite-differences-framework-tp3328p3329.html

On 2004.09.08 23:17, [hidden email] wrote:

> I'm having trouble with the finite differences framework of Quantlib.
>
> It seems that the two lines related to the boundary conditions are
> wrong:
>
>     L.setLowerBC(BoundaryCondition(BoundaryCondition::Neumann,
>         exercisingValue[1]-exercisingValue[0]));
>     L.setUpperBC(BoundaryCondition(BoundaryCondition::Neumann,
>         exercisingValue[gridPoints_-1]-exercisingValue
> [gridPoints_-2]));

Yes, the documentation is out of date. I'll try and update it when I  
find some time. But we're about to try and refactor the framework in  
the near future, and given that the time we can give to QuantLib is  
limited, it might happen that documentation is postponed until the  
framework we document is not itself outdated.

> It would be nice to provide a full working example for the finite
> differences framework.

That would be the FdAmericanOption class and all similar pricers.
They're in ql/Pricers.

Hope this helps,
                Luigi