Re: [Quantlib-dev] [Quantlib-users] Hull-White and CIR path generation

[Luigi B. <lui...@gm...>]

On Fri, 2010-10-15 at 16:50 +0100, Simon Ibbotson wrote:
> I've written some code for generating random numbers for the Gamma
> distribution, the Chi-squared distribution and the Non-Central
> Chi-Squared distribution but I need to wrap these up into a path
> generator for CIR which can then be used for Monte-Carlo pricing.
> [...] I don't want to use the InverseCumulativeRsg as inverting the
> CIR distribution is really slow compared to sampling from the
> Non-Central Chi-Squared.
> 
> So, I'm assuming that (to use the Monte Carlo framework) I need to
> create 
> 1) A CIR sequence generator.
> 2) A RNG traits template class which enables the CIR sequence generator
> class. 
> 
> Is this correct? Or is there something that I've missed?

Correct.  You can already generate numbers according to your
distribution, so just wrap them in a sequence generator and you should
be done.

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