[Quantlib-dev] Preliminary 0.9.0 tarballs - SwapRateHelper::ImpliedQuote

[Toyin A. <toy...@ho...>]

Hi all,
=20
I'm slightly confused as to how the spread is treated/used within this func=
tion.
=20
I first thought it was a swap spread, thus this spread would be simply adde=
d onto the inputted swap rate and the stripper would do it's stuff.
=20
But looking at it, it's more like a spread that is added onto every floatin=
g rate fixing within the underlying swap. This spread looks more like a bas=
is swap spread.
=20
Am I correct in assuming this?
=20
If this is the case how does one price a basis swap from the construction o=
f two floating legs back to back via this spreaded curve? I'm not really co=
ncerned about the actual QuantLib classes involved, just where does one pas=
s in the yieldcurve handles.
=20
To be a little clearer, we would now have two yieldcurves, (one without bas=
is spreads and one with) and two different yieldcurve inputs for each float=
ing leg.
=20
Leg1 will have a discounting curve (probably embedded in a discounting engi=
ne) and a fixing curve (embedded in the index).
=20
The same is true for leg2
=20
Thus if one wanted to recreate the basis swap value entered into the yieldc=
urve stripper (lets say the 10Y point), what are the input yieldcurve combi=
nations  (discounting/refixing) and (spreaded/non spreaded) for each leg?
=20
Also there is a comment of "weak implementation" stated within this method.=
..
=20
Thanks in advance,
Toy out...
=20
=20
_________________________________________________________________
The next generation of MSN Hotmail has arrived - Windows Live Hotmail
http://www.newhotmail.co.uk=