Problem with FedFund discounting curve calibration

Posted by Grison PG Pierre (External DEXIA-US) on Oct 02, 2014; 8:02pm
URL: http://quantlib.414.s1.nabble.com/Problem-with-FedFund-discounting-curve-calibration-tp15933.html

Hi,

 

I try to calibrate a USD discounting curve (so, FedFund or FF) with these instruments:

-short end: (OIS) FF = spread1 = fixe1

-long end: ( (Basis Swap or BS) FF + spread2= LIBOR 3M and (IRS) LIBOR 3M = fixe2 )

 

For the short end, everything works fine.

For the long end, my idea is to eliminate the LIBOR 3M leg in both the BS and the IRS swap to build a synthetic OIS : FF + spread2 = fixe2.

 

I have upgraded the QuantLib class OISRateHelper (and thus MakeOIS and OvernightIndexedSwap classes) in order to take into account:

-the spread “spread2”,

-the fact that both leg don’t use necessarily the same DayCounter, PaymentFrequency and so on…

Now, my OISRateHelper class takes these arguments as input and the ImpliedQuote function return the FairRate (fixe2 in the previous equation). Indeed, I prefer optimize on fixe2 value rather than on spread2 value, but I guess both can work.

 

Nonetheless, I can’t find accurate results (Murex curve is my benchmark). Is anyone interested to work on this subject/help me?

 

Regards,

 

Pierre

 

 

 

------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users