QuantLib › quantlib-users

Sobol sequences: question regarding correct usage

Classic

List

Threaded

2 messages Options

langda tyagi

Sobol sequences: question regarding correct usage

Hi,

I am using QuantLib, but my question is somewhat more conceptual in nature, and was hoping someone might have an answer.

I am valuing a derivative (european style) with 45 underlying stock prices, to be averaged over last 90 days of the term.

I am using Quasi Random nos (sobol), but not sure I am doing it right.

First I am not sure what is the dimension that I should sample the sobol for.

Is it:

a. 45 (one for each stock)

b. 90 (one for each day)

c. 4050 (45 * 90, though I suspect in this case using quasi random nos might not be suitable)

I would really appreciate your inputs on this, or if you could point me in some direction.

regards

Tyagi

------------------------------------------------------------------------------
WhatsUp Gold - Download Free Network Management Software
The most intuitive, comprehensive, and cost-effective network
management toolset available today. Delivers lowest initial
acquisition cost and overall TCO of any competing solution.
http://p.sf.net/sfu/whatsupgold-sd
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users

Kakhkhor Abdijalilov

Re: Sobol sequences: question regarding correct usage

Hi,
If each asset is driven by its own Brownian motion then the dim is
45*90. Sobol sequence is more efficient if you let lower dimensions to
do more work. To improve convergence use principal component analysis
and Brownian bridge.

Regards,
Kakhkhor Abdijalilov.

On Fri, May 6, 2011 at 12:40 AM, langda tyagi <[hidden email]> wrote:

> Hi,
>
> I am using QuantLib, but my question is somewhat more conceptual in nature,
> and was hoping someone might have an answer.
>
> I am valuing a derivative (european style) with 45 underlying stock prices,
> to be averaged over last 90 days of the term.
>
> I am using Quasi Random nos (sobol), but not sure I am doing it right.
>
> First I am not sure what is the dimension that I should sample the sobol
> for.
>
> Is it:
>
> a. 45 (one for each stock)
>
> b. 90 (one for each day)
>
> c. 4050 (45 * 90, though I suspect in this case using quasi random nos
> might not be suitable)
>
>
> I would really appreciate your inputs on this, or if you could point me in
> some direction.
>
>
> regards
>
> Tyagi
>
>
> ------------------------------------------------------------------------------
> WhatsUp Gold - Download Free Network Management Software
> The most intuitive, comprehensive, and cost-effective network
> management toolset available today. Delivers lowest initial
> acquisition cost and overall TCO of any competing solution.
> http://p.sf.net/sfu/whatsupgold-sd
> _______________________________________________
> QuantLib-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>
>