wilmott.com

Dear AllI am just wondering if you can help me on this one. I am trying to build a zero coupon yield curve for EQ and FX option pricing, say for the US. Conventionally, it is ok to use cash rate, interest rate futures (Euro$ 3m) and swap rate to obtain the corresponding zero-coupon discount facts for pricing. I am just wondering what instrument for building the zero-coupon curve should be using for treasury bond option pricing? It seems quite unreasonable to use the euro$ and swap again since they involves credits issue, should I use the discount windows and discounted T-bill rates for building the curve?Pls advisedan

Ah, no, there is no difference. You'd still use the regular swap curve to price.Regards.

Treasury Curve defines your 'Forwards'Libor curve defines your payouts.So as FDAX says.Think of it like a pseudo equity quantord = domestic rate (for option payout)rf = foreign rate (treasury yield) defines 'forwards'a bit cack-handed an explanation, I know.

Just make sure you don't use par rates, when you should be using spot rates without changin them first.Generally I almost always use the swap curve, which I convert to a spot curve, because the small credit risk embedded in LIBOR seems standard across currencies, and seems to be consistent when I price equity and FX derivatives across countries.I don't really see the advantage of using any treasury curves unless those securities are part of my trade, but I'm open to hearing why maybe I should...

I had seen in Brigo's text that for option pricing, fitting/calibrating a particular model to swaptions will make the model poorly fit to caps and vice versa .(Please correct me if I understood wrong!) .I wonder, if it is preferable to use swaptions for pricing of swap related instruments / treasury curve for captions? Please correct me if I am wrong.Thanks,asd

No asd, you did not understand the point.YOU NEVER USE TREASURY CURVEThe Lognormal Forward-libor Model(LFM) or also called Brace-Gaterak-Musiela(BGM) is a model which when you use to price caps, gives you the Black's formula for caps.The Lognormal forward Swap Model(LSM) or Jamshidian's model is yields the Black's formula for swaptions when you use it to price swaptions.But the two models are not compatible because if forward Libor are lognormal each under its proba measure, forward swap rates can be shown not to be lognormal at the same time under their proba measures.But in the paragraphs 6.9 to 6.16, they connect swaptions to the LFM. the more interesting 6.15 and 6.16. In chap 7, they calibrate the LFM to swaptions. Calibration to swaptions is obtained through an approximation on swaption prices.My advice: use the LFM but calibrated to cap ands swaptions.

Madmax, Thank you very much for your kind help and the explanation.Regards,asd

QuoteOriginally posted by: madmax YOU NEVER USE TREASURY CURVEEven for T-Bond options?