wilmott.com

Hi, need a quick and dirty way of determining the vol level for an amortising swaption from the standard swation vol surface. Any suggestions please?

Why should an amortsing swaption have a different volatility than a plain vanilla one?

Because the underlying is different from the standard swaption. Could have situation where one would be exercised and the other not for example.

Amortizing swaps have lower durations than otherwise identical vanilla swaps and vol is clearly a function of time. Pat Hagan has addressed this topic in his writings.

Can you suggest a reference please?

If you want a quick and dirty way to determine it, just take the implied volatility of a not amortized swap having the same duration as your amortised swap: ex : you want the vol of a 1m10y atm swaption constantly amortised ? = knowing that the duration of a 10y amortised is about the same as a 5y not amortised swap => take the implied vol of a 1m5y swaption.Any opinion on this method will be greatly appreciated.

Your method sounds OK, but not good enough based on my experience, As Pat addressed this issue in his paper, I tested all 3 methods and the basket method is equivalent to the matching of moneyness duration and covenxity method. so to make it simple matching of moneyness duration and covenxity will give you best result on this issue and also very easy to implement

Hi,Where can weind the document of Pat Hagan on Amortizing Swaps.Thanks.