Bootstrapping US Treasury Curve with fake zero coupon bonds
Locked
Bootstrapping US Treasury Curve with fake zero coupon bonds
 
|
Hi, I'm trying to bootstrap treasury curve with cubic spline including t-bill( short dated with no coupon) Since quantlib doesn't support zeros I add fixed rate bond helper with real price and very small coupon. During bootstrap I get an error 1st iteration: could not bootstrap the 1st instrument, maturity February 19th, 2009: root not bracketed: f[2.22045e-016,3] -> [-9.989025e+001,-4.726847e+001] >From the point of view of yield calculation price 99.89 and coupon 0.00001 makes sense. Can anyone suggest whet I'm doing wrong or it's a bug and discount bonds(price less than 100) are not handled properly? Below are the bonds used to fitting: CUSIP HASCPN FREQ CPN ISSUEDATE MATDATE BBGNAME VALUEDATE PRICE 912795K42 N 0 0 8/21/2008 2/19/2009 CB3 Govt 11/18/2008 0.11 912795L90 N 0 0 11/20/2008 5/21/2009 CB6 Govt 11/18/2008 0.74 912795S44 N 0 0 10/23/2008 10/22/2009 CB12 Govt 11/18/2008 0.99 912810PX0 Y 2 4.5 8/15/2008 5/15/2038 CT30 Govt 11/18/2008 106.9844 912828JP6 Y 2 1.5 10/31/2008 10/31/2010 CT02 Govt 11/18/2008 100.75 912828JQ4 Y 2 2.75 10/31/2008 10/31/2013 CT05 Govt 11/18/2008 102.8438 912828JR2 Y 2 3.75 11/17/2008 11/15/2018 CT10 Govt 11/18/2008 102.2656 For zeros I'm passing 100-price as input to quantlib and coupon 0.00001, Frequency.Annual, BusinessDayConvention.Following Irakli _________________________________________________________________ Get more done, have more fun, and stay more connected with Windows Mobile®. http://clk.atdmt.com/MRT/go/119642556/direct/01/ ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ QuantLib-users mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-users |
Re: Bootstrapping US Treasury Curve with fake zero coupon bonds
|
|
On Wed, 2008-11-19 at 09:38 -0500, Irakli Machabeli wrote:
> I'm trying to bootstrap treasury curve with cubic spline including t-bill( short dated with no coupon) > Since quantlib doesn't support zeros I add fixed rate bond helper with real price and very small coupon. > During bootstrap I get an error Apologies for the delay---can you post the code that triggers the problem? Luigi -- There's no sense in being precise when you don't even know what you're talking about. -- John von Neumann ------------------------------------------------------------------------------ SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada. The future of the web can't happen without you. Join us at MIX09 to help pave the way to the Next Web now. Learn more and register at http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/ _______________________________________________ QuantLib-users mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-users |
Re: Bootstrapping US Treasury Curve with fake zero coupon bonds
|
|
Luigi,
I get this error if I try to bootstrap curve from c# code 1st iteration: could not bootstrap the 1st instrument, maturity March 19th, 2009: root not bracketed: f[2.22045e-016,3] -> [-9.996249e+001,-4.730857e+001] I made sample using C++ too. Now in C++ sample I'm actually getting memory corruption error in some cases. If you exclude 1st 3 bonds in source code all tests pass smoothly (it works also from c# side). With those 3 bonds included you get an error One year zero=0.558764 % Actual/Actual (ISDA) simple compounding done fitCurve<Discount,LogLinear,IterativeBootstrap> failed fitCurve<ForwardRate,Cubic,IterativeBootstrap> 1st iteration: could not bootstrap the 3rd instrument, maturity November 19th, 2 009: root not bracketed: f[2.22045e-016,3] -> [4.166294e-001,3.803735e+001] One year zero=0.525826 % Actual/Actual (ISDA) simple compounding done fitCurve<ZeroYield,Cubic,IterativeBootstrap> One year zero=0.547035 % Actual/Actual (ISDA) simple compounding done fitCurve<ForwardRate,ConvexMonotone,IterativeBootstrap> If you do not catch exception in code debugger shows that there is memory corruption : Unhandled exception at 0x004b8ee0 in mytestsValuation-vc80-mt-sgd.exe: 0xC0000005: Access violation reading location 0xfeeefef2. so I guess "root not bracketed" is not a real issue By the way is you change frequency for 1st 3 bonds from Annual to Monthly ( which should nor really matter since coupon is really small one more test fails) Anyway attached is the sample : #include <iostream> #include <ql/quantlib.hpp> using namespace std; using namespace QuantLib; #if defined(QL_ENABLE_SESSIONS) namespace QuantLib { Integer sessionId() { return 0; } } #endif #define LENGTH(a) (sizeof(a)/sizeof(a[0])) struct BondData { Date issueDate; Date redemptionDate; Frequency frequency; Rate coupon; Real price; }; BondData bondData[] = { {Date(18,(Month)9,2008),Date(19,(Month)3,2009), Annual,0.01,99.9600}, {Date(18,(Month)12,2008),Date(18,(Month)6,2009), Annual,0.01,99.7350}, {Date(20,(Month)11,2008),Date(19,(Month)11,2009), Annual,0.01,99.5300}, {Date(1,(Month)12,2008),Date(30,(Month)11,2010), Semiannual,1.25,100.9531}, {Date(1,(Month)12,2008),Date(30,(Month)11,2013), Semiannual,2,102.3281}, {Date(17,(Month)11,2008),Date(15,(Month)11,2018), Semiannual,3.75,110.7813}, {Date(15,(Month)8,2008),Date(15,(Month)5,2038), Semiannual,4.5,130.5000} }; Date today=Date(16,(Month)12,2008);//calendar.adjust(Date::todaysDate()); template <class T, class I, template<class C> class B> void fitCurve(const I& interpolator = I(), Real tolerance = 1.0e-9) { Size bondCount = LENGTH(bondData); Calendar calendar= JointCalendar( UnitedStates(UnitedStates::Market::GovernmentBond ), UnitedKingdom(UnitedKingdom::Market::Settlement ), JointCalendarRule::JoinHolidays ); //Natural settlementDays; std::vector<boost::shared_ptr<SimpleQuote> >prices(bondCount); std::vector<Schedule> schedules(bondCount); Settings::instance().evaluationDate() = today; Natural bondSettlementDays=2; Date settlementDay= calendar.advance(today,bondSettlementDays,Days); DayCounter bondDayCounter= ActualActual(); BusinessDayConvention bondConvention= Following; Real bondRedemption=100; std::vector<boost::shared_ptr<RateHelper> > bondHelpers(bondCount ); for (Size i=0; i<bondCount ; i++) { prices[i]= boost::shared_ptr<SimpleQuote>( new SimpleQuote(bondData[i].price) ); std::vector<Rate> coupons(1, bondData[i].coupon/100.0); schedules[i] = Schedule( bondData[i].issueDate, bondData[i].redemptionDate , Period(bondData[i].frequency), calendar, bondConvention, bondConvention, DateGeneration::Backward, false); Handle<Quote> p(prices[i]); bondHelpers[i] = boost::shared_ptr<RateHelper>(new FixedRateBondHelper(p, bondSettlementDays, bondRedemption, schedules[i], coupons, bondDayCounter, bondConvention, bondRedemption, bondData[i].issueDate)); } boost::shared_ptr<YieldTermStructure> yc( new PiecewiseYieldCurve<T,I,B>(bondSettlementDays, calendar, bondHelpers, bondDayCounter)); InterestRate r= yc->zeroRate(1, Compounding::Simple ); Date maxD= yc->maxDate(); cout<< "One year zero="<<r<<endl;// <<" maxDate= "<<maxD <<endl; } int main(int, char* []) { //fitCurve<ForwardRate,Cubic,IterativeBootstrap>(); try { fitCurve<Discount,LogLinear,IterativeBootstrap>(); cout<<"done fitCurve<Discount,LogLinear,IterativeBootstrap>"; } catch( QuantLib::Error& e ) { cout<<"failed fitCurve<Discount,LogLinear,IterativeBootstrap>"<<endl<<e.what(); } cout<<endl<<endl; try { fitCurve<ForwardRate,Cubic,IterativeBootstrap>(); cout<<"done fitCurve<ForwardRate,Cubic,IterativeBootstrap>"; } catch( QuantLib::Error& e ) { cout<<"failed fitCurve<ForwardRate,Cubic,IterativeBootstrap>"<<endl<<e.what(); } cout<<endl<<endl; try { fitCurve<ZeroYield,Cubic,IterativeBootstrap>(); cout<<"done fitCurve<ZeroYield,Cubic,IterativeBootstrap>"; } catch( QuantLib::Error& e ) { cout<<"failed fitCurve<ZeroYield,Cubic,IterativeBootstrap>"<<endl<<e.what(); } cout<<endl<<endl; try { fitCurve<ForwardRate,ConvexMonotone,IterativeBootstrap>(); cout<<"done fitCurve<ForwardRate,ConvexMonotone,IterativeBootstrap>"<<endl<<endl; } catch( QuantLib::Error& e ) { cout<<"failed fitCurve<ForwardRate,ConvexMonotone,IterativeBootstrap>"<<endl<<e.what(); } cout<<endl<<endl; char c; cin>>c; } -------------------------------------------------- From: "Luigi Ballabio" <[hidden email]> Sent: Wednesday, December 17, 2008 10:15 AM To: "Irakli Machabeli" <[hidden email]> Cc: <[hidden email]> Subject: Re: [Quantlib-users] Bootstrapping US Treasury Curve with fake zero coupon bonds > On Wed, 2008-11-19 at 09:38 -0500, Irakli Machabeli wrote: >> I'm trying to bootstrap treasury curve with cubic spline including >> t-bill( short dated with no coupon) >> Since quantlib doesn't support zeros I add fixed rate bond helper with >> real price and very small coupon. >> During bootstrap I get an error > > Apologies for the delay---can you post the code that triggers the > problem? > > Luigi > > > -- > > There's no sense in being precise when you don't even know what you're > talking about. > -- John von Neumann > > > ------------------------------------------------------------------------------ SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada. The future of the web can't happen without you. Join us at MIX09 to help pave the way to the Next Web now. Learn more and register at http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/ _______________________________________________ QuantLib-users mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-users |
Re: Bootstrapping US Treasury Curve with fake zero coupon bonds
|
|
Hi,
I am also having similar issues bootstrapping a treasury curve with QuantLib. Has there been any resolution to this thread? Any help would be appreciated, Peter
|
Re: Bootstrapping US Treasury Curve with fake zero coupon bonds
|
|
On Tue, 2009-10-20 at 11:51 -0700, pdrubetskoy wrote:
> I am also having similar issues bootstrapping a treasury curve with > QuantLib. Has there been any resolution to this thread? Not that I know of. However, my guess is that unconstrained Cubic interpolation oscillates too much and the bootstrapper cannot find a solution; other interpolations work. Would you have time to debug the problem? Luigi -- I'd never join any club that would have the likes of me as a member. -- Groucho Marx ------------------------------------------------------------------------------ Come build with us! The BlackBerry(R) Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9 - 12, 2009. Register now! http://p.sf.net/sfu/devconference _______________________________________________ QuantLib-users mailing list [hidden email] https://lists.sourceforge.net/lists/listinfo/quantlib-users |
Re: Bootstrapping US Treasury Curve with fake zero coupon bonds
|
|
Thanks, Luigi.
This particular problem was actually caused by me passing 0 for bond redemptions by mistake, sorry!
|
«
Return to quantlib-users
|
1 view|%1 views
| Free forum by Nabble | Edit this page |