wilmott.com

Hello,I am looking at a TARN structure and wondering whether this can safely be priced using a single-factor Monte Carlo approach. The point I am unsure about is that cash flows are not Markovian in that each payment depends on the full history of payments up to that point.I would be delighted to receive any suggestions or references regarding TARN pricing in general or the choice between one- and two- factor models in particular.Thanks!

correlation and skew are both important. don't use one factor.

Thanks for that mj.

Sorry MJ,I am surprised by your answer: I agree that skew (and more generally volatility short term dynamics) is important.But why do you say that decorrelation matters ? Cheers.

I'm new to Exotic interest rate products, would anyone be able to explain the structure of a TARN and how its typically hedged?Many Thanks-Raph

TARN = TArget Redemption NoteA typical structure may be:Semi annual paying note - mat 10 years, target 12%.1st Year pay 8%2nd year onwards - if 6m Libor < 3% at fix date, pay 3% else pay zero.Once cumulative life to date coupon hits target note redeems. If this never happens then redeem at maturity (and may or may not make up coupon to target level at that point).Typically barrier will be set so forwards would imply note redeems quickly - but only just (say rate needs to move up 25bps before barrier is hit). Buyer is selling option on 6m Libor which can look very binary.Rationale is buyer gets above market coupon in exchange for selling issuer a big option.I presume MJ means correlation between libors as evolution of term structure matters.

Hi,What kind oh model can be used to valuate this product. Is Hull & White model calibrate on swaption volatility a model to price this product ?What is the market pratice for the pricing of this kind of product ?Regards