Black-Scholes Process required in MCHimalayanEngine
Locked
Black-Scholes Process required in MCHimalayanEngine
 
|
Hi,
apart from the determination of the risk free rate to discount the final option payment, is there any reason why the MCHimalayanEngine resp. its StochasticProcessArray demands the usage of a (Generalized)BlackScholesProcess-derived class? I would like to use a GeometricBrownianMotionProcess to model the underlying indices... Rgds, Andreas ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
On 26 November 2010 17:38, Andreas Spengler <[hidden email]> wrote:
> StochasticProcessArray demands the usage of a > (Generalized)BlackScholesProcess-derived class? > > I would like to use a GeometricBrownianMotionProcess to model the > underlying indices... Perhaps you could feed GenBSProc with the following: - flat interest rates curve with a rate that suits you (mu + 0.5*sigma^2) - flat dividend yield curve with a rate equal to 0 - BlackConstantVol initiated with your sigma M. ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
Hi,
Am 27.11.2010 15:35, schrieb Marcin Pawlik: > Perhaps you could feed GenBSProc with the following: > - flat interest rates curve with a rate that suits you (mu + 0.5*sigma^2) > - flat dividend yield curve with a rate equal to 0 > - BlackConstantVol initiated with your sigma in the meantime I looked at the code. I would propose to pull up the riskFreeRate member to StochasticProcess and add a corresponding constructor with an empty default parameter. One could then check in all necessary situations whether riskFreeRate contains a valid object... ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
In reply to this post by Marcin Pawlik
Hi Marcin,
Am 27.11.2010 15:35, schrieb Marcin Pawlik: > Perhaps you could feed GenBSProc with the following: > - flat interest rates curve with a rate that suits you (mu + 0.5*sigma^2) > - flat dividend yield curve with a rate equal to 0 > - BlackConstantVol initiated with your sigma How would that make a GenBSProc represent a _Geometric_BrownianMotian? Rgds, Andreas ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
On 27 November 2010 19:18, Andreas Spengler <[hidden email]> wrote:
> Hi Marcin, > > Am 27.11.2010 15:35, schrieb Marcin Pawlik: > >> Perhaps you could feed GenBSProc with the following: >> - flat interest rates curve with a rate that suits you (mu + 0.5*sigma^2) >> - flat dividend yield curve with a rate equal to 0 >> - BlackConstantVol initiated with your sigma > > How would that make a GenBSProc represent a _Geometric_BrownianMotian? I used GenBSProc as short for GeneralizedBlackScholesProcess. I understood that you want to use in your simulation something as simple as GeometricBrownianMotion where the simulated trajectories are determined by constant parameters (mu, sigma). Unfortunately the MC himalaya option engine uses StochasticProcessArray consisting of n GeneralizedBlackScholesProcesses and those are not as simple as the thing you wanted to use (i.e. GeneralizedBlackScholesProcess). What I'm trying to propose is to emulate GeometricBrownianMotion (GBM for short) with an instance of GeneralizedBlackScholesProcess (GBSP for short). Since you cannot instantiate GBSP with constants as you're doing it with GBM, you may try to instantiate GBSP using flat yield curve (FlatForwardCurve class) and const vol (BlackConstantVol). Such approach would result in GBSP with constant drift coefficient (just like in GBM) and constant volatility (again just like in GBM). Is it what you're looking for? M. ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
Am 27.11.2010 19:36, schrieb Marcin Pawlik:
> What I'm trying to propose is to emulate GeometricBrownianMotion (GBM > for short) with an instance of GeneralizedBlackScholesProcess (GBSP > for short). Since you cannot instantiate GBSP with constants as you're > doing it with GBM The GBM has no constant coefficients, since it's coefficients are dependant on S also. That's why I want to use a GBM; it represents a different SDE. I can adapt the MCHimalayaEngine to get the riskFreeRate from somewhere else, but my original question was, whether there is anywhere deep in the MC code that also demands the StochasticProcess derived class to actually be a (Generalized)BlackScholesProcess... Rgds, Andreas ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
On Nov 27, 2010, at 8:39 PM, Andreas Spengler wrote: > I can adapt the MCHimalayaEngine to get the riskFreeRate from > somewhere > else, but my original question was, whether there is anywhere deep in > the MC code that also demands the StochasticProcess derived class to > actually be a (Generalized)BlackScholesProcess... No, no deep reason as far as I know. I'd change the constructor to take a generic process and a handle to a discount curve, and use the first for path generator ad the second for discounting the payoff. I'd be grateful if you contributed the resulting code. Luigi ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
Am 27.11.2010 22:22, schrieb Luigi Ballabio:
> I'd change the constructor to take a generic process and a handle to > a discount curve, and use the first for path generator ad the second > for discounting the payoff. I'd be grateful if you contributed the > resulting code. Gladly. However, instead of changing the interface for the affected Engine classes, I would rather, as stated in my other mail, propose to pull up the riskFreeRate member of GeneralizedBlackScholesProcess to the StochasticProcess class and add another parameter (empty by default) in e.g. GeometricBrownianMotionProcess' or StochasticProcessArray's constructor. One could then check in all relevant situations, whether riskFreeRate contains a valid object... ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
On Sat, 2010-11-27 at 23:41 +0100, Andreas Spengler wrote:
> However, instead of changing the interface for the affected > Engine classes, I would rather, as stated in my other mail, propose to > pull up the riskFreeRate member of GeneralizedBlackScholesProcess to the > StochasticProcess class and add another parameter (empty by default) in > e.g. GeometricBrownianMotionProcess' or StochasticProcessArray's > constructor. Hmm. I'm not sure I would do that. For one thing, even when using the Black-Scholes model, one might want to use a different curve for the discount; asking the process for the risk-free rate would prevent that. Furthermore, I'm not sure that a riskFreeRate method belongs to the generic StochasticProcess interface. Luigi -- Blessed is the man who, having nothing to say, abstains from giving wordy evidence of the fact. -- George Eliot ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
Am 29.11.2010 17:16, schrieb Luigi Ballabio:
> Hmm. I'm not sure I would do that. For one thing, even when using the > Black-Scholes model, one might want to use a different curve for the > discount; asking the process for the risk-free rate would prevent that. > Furthermore, I'm not sure that a riskFreeRate method belongs to the > generic StochasticProcess interface. I second that, however I found another problem with using the GeometricBrownianMotionProcess as is: it inherits an otherwise empty time(const Date&) method from StochasticProcess which simply QL_FAILs... ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
Re: Black-Scholes Process required in MCHimalayanEngine
|
|
On Mon, 2010-11-29 at 20:15 +0100, Andreas Spengler wrote:
> Am 29.11.2010 17:16, schrieb Luigi Ballabio: > > > Hmm. I'm not sure I would do that. For one thing, even when using the > > Black-Scholes model, one might want to use a different curve for the > > discount; asking the process for the risk-free rate would prevent that. > > Furthermore, I'm not sure that a riskFreeRate method belongs to the > > generic StochasticProcess interface. > > I second that, however I found another problem with using the > GeometricBrownianMotionProcess as is: it inherits an otherwise empty > time(const Date&) method from StochasticProcess which simply QL_FAILs... Right---the date/time conversion. Hmm. For the time being, you might have the process take a DayCounter and a referenceDate, and implement time() in terms of those. Sigh. We'll have to rethink this stuff one day... Luigi -- Better to have an approximate answer to the right question than a precise answer to the wrong question. -- John Tukey as quoted by John Chambers ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ QuantLib-dev mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-dev |
«
Return to quantlib-dev
|
1 view|%1 views
| Free forum by Nabble | Edit this page |