Serving the Quantitative Finance Community

 
User avatar
kulmoedee
Topic Author
Posts: 0
Joined: October 28th, 2003, 3:10 am

*** CDS Unwind Value ****

September 14th, 2004, 7:32 pm

I need a little CDS primer. Let’s say I buy protection at 100bps for 5yrs on $10mm notional. A year passes and CDS is quoted 100bps wider (e.g. the broker quotes that I can sell protection at 200bps). How is the CDS pay-off calculated if I choose to unwind the trade? More specifically, is the unwind value simply 100bps over 4yrs x $10mm discounted back at each period by the risky rate or is there more to the computation. Thx for your help.
Last edited by kulmoedee on September 13th, 2004, 10:00 pm, edited 1 time in total.
 
User avatar
BeZen
Posts: 0
Joined: May 14th, 2004, 7:46 am

*** CDS Unwind Value ****

September 15th, 2004, 7:06 am

You will unwind it at the 4Y bid level (say x bp).Then the pay-off is : $10mm*(x-100)*"remaining duration" - accrued value of your current CDSTo compute the remaining duration you need to add the risk free curve + the CDS spread curveYou can also do it with the CDSW function in Bloomberg ...BeZen
Last edited by BeZen on September 14th, 2004, 10:00 pm, edited 1 time in total.
 
User avatar
kulmoedee
Topic Author
Posts: 0
Joined: October 28th, 2003, 3:10 am

*** CDS Unwind Value ****

September 15th, 2004, 8:00 pm

BeZen, thx for your response. If you don't mind, could you tell me how you specifically compute the "remaining duration"?Could someone please elaborate further on how the CDS unwind value is computed. More specifically, I guess my question really is how does the value of unwinding a CDS differ from that of a shorting and covering a cash bond.? For example, if I shorted a cash bond at 100bps over Libor and it subsequently widened out to 200bps would my payoff be exactly the same if i bot protection at 100bps and unwinded it at 200bps (excluding supply/demand pricing differences in the 2 different markets - I'm interested in the math behind the CDS unwind value). Thx.
 
User avatar
pickles
Posts: 0
Joined: July 14th, 2002, 3:00 am

*** CDS Unwind Value ****

September 16th, 2004, 3:50 am

risky duration or dv01 is;df(i) *dcf(i) * probsur(i)where df is riskless discount factordcf is the day count fractionand probsur(i) is the probability of surviving to the cashflow timeIn the example you give there will be a difference in price if there are more than one bonds out there to deliver into the cds. If there are then the cds has a cheapest to deliver option. CDS spread will therefore be higher since you would pay more for this option.
 
User avatar
Herbie
Posts: 0
Joined: May 5th, 2003, 2:17 pm

*** CDS Unwind Value ****

September 16th, 2004, 1:05 pm

try this approximation... doesn't take into account term structure, but pretty good otherwise:pv = (CS - X) . (1 - exp( - (h + r).T)) / (h+r)where r = in (1+ swp rate for maturity T)h = CS / 1-RRT = CDS maturityCS = mkt credit spd for your maturity CDS (i.e. interpolate on the term structure)X = contractual credit spreadThe term after (CS-X) is roughly the risky duration, i.e. the number of years that you accrue CS - X for.Really though you need to do this calculation and calibrate multiple points of the credit spread by asking 'what should the hazard rate be such that the calibration cds has zero value'.