[R-SIG-Finance] Calibration of Heston Model in R

[Enrico Schumann]

On Sun, 13 Jul 2014, Shivam <shivamsingh at gmail.com> writes:

> Hi All,
>
> It is a very basic question, in the sense that I need to start from
> scratch. I need to know what are the resources available in R to calibrate
> the Heston model.
>
> A lot of help is available for MATLAB, but I have been working with R for a
> while now and dont want to migrate.
>
> I know the basic idea of the calibration and the theory around the model,
> but am at a loss as to how do I really go about the calibration in R.
>
> Any inputs would be of great help.
>
> Thanks in advance,
> Shivam
>

It is some time that I have worked with the Heston model (and I have
never used it since, and it was with MATLAB), but a few pointers.

In terms of computational tools you will need two things:

1) A pricing mechanism (there are several)

   There is an implementation of a pricing method in the NMOF package
   [1], but for calibration you may need something faster. The pricing
   method we used is described in detail in chapter 15 of Gilli et al
   [2].  I had started a translation into R in the NMOF manual [3], but
   it is not complete.
 
2) An optimisation algorithm (there are several)

   Have a look at the Optimization Task View [4].  Since the problem is
   only mildly constrained, I would suggest Differential Evolution.
   (Classical -- derivatives-based -- methods did not work very well for
   me.)  See package DEoptim or function DEopt in package NMOF.

As I said, just a few pointers.

Regards,
        Enrico


[1] Disclosure: I am the package author.
[2] http://enricoschumann.net/NMOF.htm#Book
[3] http://enricoschumann.net/NMOF.htm#NMOFmanual
[4] http://cran.r-project.org/web/views/Optimization.html


-- 
Enrico Schumann
Lucerne, Switzerland
http://enricoschumann.net