|
From: Ioannis R. <qua...@de...> - 2026-03-09 09:58:38
|
As I and Peter wrote earlier, you do not need native OIS support to price a Bermudan OIS option. You can use the QuantLib type VanillaSwaption. It references an underlying ibor swap with a fixed maturity date. If you need an OIS underlying, you can equivalently use a VanillaSwap that references a custom ibor of which the tenor equals the period of the fixed leg of the OIS. Then you will get the exact same price. On 3/9/2026 4:18 AM, Rich Amaya via QuantLib-users wrote: > Hi Peter and Daniel, > > Thanks for the clarification Peter. I’ve opened a GitHub issue to > track adding native OIS support to the FD Bermudan swaption engines. > > Best, > Rich > >> On Mar 8, 2026, at 1:23 PM, Peter Caspers <pca...@gm...> wrote: >> >> Hi Rich and Daniel, >> >> Apologies, I misunderstood the trade terms - I thought we were >> talking about a swaption exercising into an underlying swap with >> constant maturity measured from the respective exercise date! >> >> Thanks for clarifying. >> >> Best >> Peter >> >> On Sun, 8 Mar 2026 at 18:25, Rich Amaya <ric...@ma...> wrote: >> >> Hi Daniel, >> >> As Peter says, QuantLib doesn’t have a Bermudan swaption with an >> OIS underlying. However, you can use a VanillaSwap as a proxy for >> the underlying and price it with the existing Bermudan >> infrastructure. >> >> The existing QuantLib examples are actually the co-terminal case >> you need - a single swap with a fixed maturity date, with >> exercise dates at each fixed coupon accrual start. When exercised >> at a later date, only the remaining coupons contribute to the >> payoff, so the effective swap tenor is shorter while the maturity >> date stays the same. See >> Examples/BermudanSwaption/BermudanSwaption.cpp (line 248 onward) >> and test-suite/bermudanswaption.cpp. >> >> The setup is: create a VanillaSwap with your fixed maturity date, >> build a BermudanExercise from the fixed leg accrual start dates >> (your 5 exercise dates), and price with FdHullWhiteSwaptionEngine. >> >> Under a one-factor Hull-White model with a single forwarding >> curve, this VanillaSwap proxy is exact - the compound of daily >> overnight forwards equals the period forward rate on the same >> curve, so VanillaSwap and OIS floating legs produce identical NPVs. >> >> *_References_* >> - Brigo & Mercurio, “Interest Rate Model — Theory and Practice” >> (Ch. 3 for HW model, tree-based pricing) >> - Hull, “Options, Futures, and Other Derivatives,” 11th ed. (Ch. >> 31-32 for short-rate models and trinomial trees) >> >> I hope that helps! >> >> Best, >> Rich >> >> >>> On Mar 8, 2026, at 6:53 AM, Daniel Lobo >>> <dan...@gm...> wrote: >>> >>> Hi Peter, >>> >>> Thanks for your response. >>> >>> Do you know if there is any online resource available to get >>> guidance >>> on this pricing and if QuantLib has any pipeline to implement this >>> Instrument? >>> >>> BR, >>> >>> On Sun, 8 Mar 2026 at 00:15, Peter Caspers >>> <pca...@gm...> wrote: >>>> >>>> Hi Daniel, >>>> >>>> I don't think we have this variant of a Bermudan swaption in >>>> QuantLib. >>>> >>>> Best >>>> Peter >>>> >>>> >>>> On Sat, 7 Mar 2026 at 09:55, Daniel Lobo >>>> <dan...@gm...> wrote: >>>>> >>>>> Hi, >>>>> >>>>> I need to price a Bermudian type swaption which has multiple >>>>> maturity >>>>> dates (total number 5) however underlying swap is OIS, which has a >>>>> fixed maturity date. >>>>> >>>>> I found some swaption examples, where underlying swap has only >>>>> fixed >>>>> maturity term, not fixed maturity date. Therefore in my case, >>>>> at eah >>>>> exercise date, underlying swap's maturity term is lesser. >>>>> >>>>> Can you please point if there any such quantlib implementation or >>>>> example to price such swaption based on hull white model? >>>>> >>>>> >>>>> _______________________________________________ >>>>> QuantLib-users mailing list >>>>> Qua...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/quantlib-users >>> >>> >>> _______________________________________________ >>> QuantLib-users mailing list >>> Qua...@li... >>> https://lists.sourceforge.net/lists/listinfo/quantlib-users >> > > > > _______________________________________________ > QuantLib-users mailing list > Qua...@li... > https://lists.sourceforge.net/lists/listinfo/quantlib-users -- This email has been checked for viruses by Avast antivirus software. www.avast.com |