Bates process

[Lapin]

Hi,

I am wondering why there is no Bates process derived from Heston process?
Does it make sense to refactor it in order to have a Bates process?

Cheers

[Luigi Ballabio]

On Feb 22, 2008, at 10:50 AM, Yomi wrote:
> I am wondering why there is no Bates process derived from Heston
> process?
> Does it make sense to refactor it in order to have a Bates process?

Yes, probably.  Klaus, are you reading this? What do you think?

Luigi


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Klaus Spanderen-2]

Hi

good point, I've just committed a BatesProcess class to the QuantLib CVS. See

http://quantlib.svn.sourceforge.net/viewvc/*checkout*/quantlib/trunk/QuantLib/ql/processes/batesprocess.cpp
http://quantlib.svn.sourceforge.net/viewvc/*checkout*/quantlib/trunk/QuantLib/ql/processes/batesprocess.hpp

Monte-Carlo Simulation works in the same way as for the HestonProcess (indeed
you might use the BatesProcess as a substitute for the HestonProcess). Please
see in test-suite/batesmodel.cpp

http://quantlib.svn.sourceforge.net/viewvc/*checkout*/quantlib/trunk/QuantLib/test-suite/batesmodel.cpp

the test case testAnalyticVsMCPricing().

hope that fulfills your needs

 Klaus

On Friday 22 February 2008 10:50:09 Yomi wrote:
> Hi,
>
> I am wondering why there is no Bates process derived from Heston process?
> Does it make sense to refactor it in order to have a Bates process?
>
> Cheers



--
Klaus Spanderen
Ludwig Erhard Str. 12
48734 Reken (Germany)
EMail: [hidden email] (remove NOSPAM from the address)

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Lapin]

Klaus,

 

I have checked the behavior of the Bates process within a MC engine I have done for different options.

I have compared the value of a 1M HSI vanilla call 90% strike with the market value and its Closed formula value after calibration.

You can see the results on the following graph:

 

It looks good to me.

I was wondering if it would not make sense in term of calculation speed to move the Normal distribution and Poisson distribution as part of the process it self? I know it does not make sense in term of abstraction, but having to create the object of each step could be coslty in term of speed.

 

What are your thougths?

Cheers

 

2008/2/23, Klaus Spanderen <[hidden email]>:

Hi

good point, I've just committed a BatesProcess class to the QuantLib CVS. See

http://quantlib.svn.sourceforge.net/viewvc/*checkout*/quantlib/trunk/QuantLib/ql/processes/batesprocess.cpp
http://quantlib.svn.sourceforge.net/viewvc/*checkout*/quantlib/trunk/QuantLib/ql/processes/batesprocess.hpp

Monte-Carlo Simulation works in the same way as for the HestonProcess (indeed
you might use the BatesProcess as a substitute for the HestonProcess). Please
see in test-suite/batesmodel.cpp

http://quantlib.svn.sourceforge.net/viewvc/*checkout*/quantlib/trunk/QuantLib/test-suite/batesmodel.cpp

the test case testAnalyticVsMCPricing().

hope that fulfills your needs

Klaus

On Friday 22 February 2008 10:50:09 Yomi wrote:
> Hi,
>
> I am wondering why there is no Bates process derived from Heston process?
> Does it make sense to refactor it in order to have a Bates process?
>
> Cheers



--
Klaus Spanderen
Ludwig Erhard Str. 12
48734 Reken (Germany)
EMail: [hidden email] (remove NOSPAM from the address)

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Klaus Spanderen-2]

Hi

yes, I'm going to define the Normal distribution as a part of the process. The
Poisson distribution depends on the evolve parameter "dt". Therefore IMO this
object has to be part of the evolve method.

cheers

On Monday 25 February 2008 11:33:33 you wrote:
> I was wondering if it would not make sense in term of calculation speed to
> move the Normal distribution and Poisson distribution as part of the
> process it self? I know it does not make sense in term of abstraction, but
> having to create the object of each step could be coslty in term of speed.
>
> What are your thougths?
> Cheers
>


--
Klaus Spanderen
Ludwig Erhard Str. 12
48734 Reken (Germany)
EMail: [hidden email] (remove NOSPAM from the address)

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users