The G2 lattice model
Hello,
My goal is to validate a G2 model Tree by first pricing vanilla instruments.
Concerning the pricing of swaptions using G2 analytical formula's and pricing numerically in a G2-2 dimensional tree (x,y,t) taken from QuantLib8.0, the gap between prices (analytic - numeric) is around 10-15 bp in average on a set of coterminal ATM swaptions 10Y.
With a HullWhite 1 Factor Tree, the gap is very small (a few bp, depending on the time grid).
I added a method for calibrating the varphi(t) function numerically in the tree (as the HullWhite Tree) to fit exactly the initial ZC curve, but this does not lead to better results for option pricing.
The parameters for computing the swaption analytical formula were chosen to be very accurate on this part.
Does anyone have experimented the same kind of problem and have an explanation for this ?
Thanks,
--------------------------------------
Dexia Bank disclaimer:
http://www.dexia.be/maildisclaimer.htm
--------------------------------------
|
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Locked
