Problems in Bermudan Swaption

[andrea.odetti-2]

I have a problem using BermudanSwaption pricer:

I have a Receiver Bermudan Swaption with a very big strike (15% for
instance)
The swap starts today.
And I can exercise the swaption in all the reset dates of the floating leg
(including the first one, ie today)

I think that the swap and the swaption should have the same value, but when
I price them, the swaption has a bigger value than he swap (this is not a
numerical difference).

You can test it in the example of 0.3.0 and 0.3.1 versions.

Open the file BermudanSwaption.cpp

1) on line 163 change
        bool payFixedRate = false;
2) on line 182 change the strike of the otmSwap to a big value: 0.15 (so I
will exercise always in the tree)
3) on line 306 insert a line to look at the swap value
        std::cout << "HW swap:  " << otmSwap->NPV() << std::endl;
4) run it

HW price (and BK too) of the bermudan is 453.648
Swap value is 411.316

and I think they should be the same

... may be I do something wrong...

bye

andrea


----------------------------------------------------------------------------------------------------------------------------------------
La presente comunicazione è destinata esclusivamente al soggetto indicato
più sopra quale destinatario o ad eventuali altri soggetti autorizzati a
riceverla. Essa contiene informazioni strettamente confidenziali e
riservate, la cui comunicazione o diffusione a terzi è proibita, salvo che
non sia stata espressamente autorizzata.
Se avete ricevuto questa comunicazione per errore, Vi preghiamo di darne
immediata comunicazione al mittente e di cancellarne ogni evidenza dai
Vostri supporti.
----------------------------------------------------------------------------------------------------------------------------------------

[Sadruddin Rejeb-3]

Hi Andrea,
Yep, there is indeed something wrong and I'm going to look into it in the next
few days. I will notice the list when the fix is commited into CVS.
Thanks for noticing this,
Sad

On Wednesday 20 November 2002 12:03, [hidden email] wrote:

>
> and I think they should be the same
>
> ... may be I do something wrong...
>
> bye
>
> -------------------------------------------------------
> This sf.net email is sponsored by: To learn the basics of securing
> your web site with SSL, click here to get a FREE TRIAL of a Thawte
> Server Certificate: http://www.gothawte.com/rd524.html
>                                                
> Quantlib-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Sadruddin Rejeb-4]

Hi Andrea,
Yep, there is indeed something wrong and I'm going to look into it in the next
few days. I will notice the list when the fix is commited into CVS.
Thanks for noticing this,
Sad

On Wednesday 20 November 2002 12:03, [hidden email] wrote:

>
> and I think they should be the same
>
> ... may be I do something wrong...
>
> bye
>
> andrea
> -------------------------------------------------------
> This sf.net email is sponsored by: To learn the basics of securing
> your web site with SSL, click here to get a FREE TRIAL of a Thawte
> Server Certificate: http://www.gothawte.com/rd524.html
> _______________________________________________
> Quantlib-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/quantlib-users

[Marco Marchioro-2]

Hi,
I'm no optimization guru, but I found the following paper

http://csep1.phy.ornl.gov/mo/node11.html

which advices to use alpha=1.e-4 and beta=0.9 for Armijo and Goldstein
criteria.
At the moment QuantLib has default values alpha=0.5 and beta=0.65.
Should we change these default values?

Did anybody ever tested this algorithm?

Marco

[Nicolas Di Césaré]

Le ven 22/11/2002 à 11:44, Marco Marchioro a écrit :

> Hi,
> I'm no optimization guru, but I found the following paper
>
> http://csep1.phy.ornl.gov/mo/node11.html
>
> which advices to use alpha=1.e-4 and beta=0.9 for Armijo and Goldstein
> criteria.
> At the moment QuantLib has default values alpha=0.5 and beta=0.65.
> Should we change these default values?
>
> Did anybody ever tested this algorithm?
>

Hi Marco,

I have extensively use this algorithm. I don't remember where I have
found that default values. I will try to find more informations.
The only thing, I remember is that beta > 0.5.

I will have a look the article next week.

Nicolas


--
Nicolas Di Césaré <[hidden email]>

[Nicolas Di Césaré]

Le ven 22/11/2002 à 19:57, Nicolas Di Césaré a écrit :

> Le ven 22/11/2002 à 11:44, Marco Marchioro a écrit :
> > Hi,
> > I'm no optimization guru, but I found the following paper
> >
> > http://csep1.phy.ornl.gov/mo/node11.html
> >
> > which advices to use alpha=1.e-4 and beta=0.9 for Armijo and Goldstein
> > criteria.
> > At the moment QuantLib has default values alpha=0.5 and beta=0.65.
> > Should we change these default values?
> >
> > Did anybody ever tested this algorithm?
> >
>
> Hi Marco,
>
> I have extensively use this algorithm. I don't remember where I have
> found that default values. I will try to find more informations.
> The only thing, I remember is that beta > 0.5.
>
> I will have a look the article next week.
>

Hi Marco,

There is an error in the default value of alpha. I have probably done a
copy/paste error (alpha=0.05!). The theoric recommandations for alpha
and beta are the following :  0 < alpha < 0.5 < beta < 1.

Thus, the both (corrected ;-) couple of values are admissible!

We are dealing with numerical algorithm, so to know which one is
better depend on the smoothness of the minimized function.

Do you think that a curve fitting example could be helpfull in the
example section?



--
Nicolas Di Césaré <[hidden email]>

[Ferdinando Ametrano-3]

At 11:25 PM 11/28/2002 +0100, Nicolas Di Césaré wrote:
>Do you think that a curve fitting example could be helpfull in the
>example section?
Good examples are always helpful ;-)
Always.


------------
ciao --Nando