Bootstrap yield curve with QLNet / Quantlib

Question

I am trying to grasp QLNet (C# version of Quantlib, all the functions of Quantlib have the same name and work the same way, so if you just know Quantlib, you can still help me), especially for pricing bonds. So I dug into the official examples, downloaded in the same time than the QLNet dll. The example called "Bonds" prices three bond types : zero coupon, fixed coupon rate, and floating coupon rate.

For the two last types, this example use two different yield curve : - The first one called discountingTermStructure for the fixed rate bond, made up of zero coupon bonds for the short end and of bonds (I'm not sure here what these simple bonds are) for the long end - The second one called forecastingTermStructure for the floating rate bond, made up of deposit for the short end and of swaps for the long end

My first question is why do they use different yield curve?

I am also trying to return a yield from a bootstrapped yield curve. For example, if I build the yield curve with 3 ZC bonds of 3 months, 6 months and 1 year maturities, I want to get the interpolated yield at 9 months. How could I do that?

Many thanks for your help

Answer 1

While @Baruch Youssin answers correctly in the general sense, the first part of his answer isn't what happened in the example code.

While QLNet is a port of QuantLib, it's not a direct port. Your quoted example doesn't show up in QLNet. The example in QuantLib was written in a very complicated way, in fact it's a simple example.

discountingTermStructure is simply a variable in the code. I can name it as something else. It links to bondDiscountingTermStructure, a yield-curve derived by today's market quotes. The yield-curve is derived by zero-coupon-rates and fixed-rate bonds.

     // Adding the ZC bonds to the curve for the short end
     bondInstruments.push_back(zc3m);
     bondInstruments.push_back(zc6m);
     bondInstruments.push_back(zc1y);

     // Adding the Fixed rate bonds to the curve for the long end
     for (Size i=0; i<numberOfBonds; i++) {
         bondInstruments.push_back(bondsHelpers[i]);
     } 

     boost::shared_ptr<YieldTermStructure> bondDiscountingTermStructure(
             new PiecewiseYieldCurve<Discount,LogLinear>(
                     settlementDate, bondInstruments,
                     termStructureDayCounter,
                     tolerance));

This curve can be used for discounting in bond pricing.

     boost::shared_ptr<PricingEngine> bondEngine(
             new DiscountingBondEngine(discountingTermStructure));

DiscountingBondEngine would ask the underlying fixed-rate bond for the cash-flow amount. This amount would need to be discounted. The discount rate is supplied by the discountingTermStructure yield-curve.

Next, we'd want to price a floating-rate bond. It requires a forward curve. In particular, we need a forward curve of 3M USD LIBOR because the bond links to the 3M LIBOR index.

    FloatingRateBond floatingRateBond(
             settlementDays,
             faceAmount,
             floatingBondSchedule,
             libor3m,
             Actual360(),
             ModifiedFollowing,
             Natural(2),
             // Gearings
             std::vector<Real>(1, 1.0),
             // Spreads
             std::vector<Rate>(1, 0.001),
             // Caps
             std::vector<Rate>(),
             // Floors
             std::vector<Rate>(),
             // Fixing in arrears
             true,
             Real(100.0),
             Date(21, October, 2005));

This forward curve bootstrapped by:

     // A depo-swap curve
     std::vector<boost::shared_ptr<RateHelper> > depoSwapInstruments;
     depoSwapInstruments.push_back(d1w);
     depoSwapInstruments.push_back(d1m);
     depoSwapInstruments.push_back(d3m);
     depoSwapInstruments.push_back(d6m);
     depoSwapInstruments.push_back(d9m);
     depoSwapInstruments.push_back(d1y);
     depoSwapInstruments.push_back(s2y);
     depoSwapInstruments.push_back(s3y);
     depoSwapInstruments.push_back(s5y);
     depoSwapInstruments.push_back(s10y);
     depoSwapInstruments.push_back(s15y);
     boost::shared_ptr<YieldTermStructure> depoSwapTermStructure(
             new PiecewiseYieldCurve<Discount,LogLinear>(
                     settlementDate, depoSwapInstruments,
                     termStructureDayCounter,
                     tolerance));

The relationship between zero rates and swap rates will tell us the forward rates.

Answer 2

I do not yet know QuantLib but one question is general and easy to answer:

My first question is why do they use different yield curve?

These two curves differ by risk levels inherent in them - the credit spreads over the risk-free yield curve (e.g., the OIS curve).

The discounting curve, discountingTermStructure, embeds the risk that this particular bond will not pay on time. Alternatively, it can be the risk-free curve if the default risk of the bond is taken care of otherwise and is not embedded in its price.

The forecast curve, forecastingTermStructure, embeds the risk of the underlying floating rate. Suppose the payments of your floating rate bond are linked to USD Libor; it is the rate at which large banks in London exchange USD deposits/loans. This rate includes the premium for the risk that borrowing bank will not pay on time.

I have also found the answer to the second question:

I want to get the interpolated yield at 9 months. How could I do that?

According to Cogito Learning QuantLib: Yield Curves, the method YieldTermStructure::zeroRate returns the zero-coupon rate for any given moment.