Basket Credit Default Swap(Correlation)

http://www.financeocean.org/finance_glossary/definition/511-basket-credit-default-swap

1. A basket credit default swap is a credit derivative contract that pays off when any of the multiple reference entities default.
   
   The swap contract defines the number of defaults after which the payoff is generated.
   
   Classifications of basket CDS's include: first-to-default CDS, second-to-default CDS or more generally nth-to-default CDS.
   

http://quantlabs.net/labs/home/896-quant-analytics-default-correlation-in-cdo-or-basket-cds

 

Quant analytics: Default correlation in CDO or basket CDS

Default correlation on basket CDO or basket credit default swap.

An example:

A reference portfolio  of 10 bonds with the same probability 1% chance of default.

Correlation 0.0 1.0

iid (%) perfect

Senior P(All default) 0 1

CDO or Basket CDS 10th 1 1

9th 1 1

8 1 1

7 1 1

6 1 1

5 1 1

4 1 1

3 1 1

2nd 1 1

1st 1 1

P[no default] 90.4% 99%

Junior P[>1 default] 9.6% 1.0%

P is probability.

The 1st is the junior tranche and the 10th would the senior tranche. The 1st could be the 1st to default tranche and the 10th would be the 10th level level to default tranche. Under iid, it would the reference portfolio. You could have a low corelation scenario above (iid) which independent. The other scenario is going to other extreme where is the perfect default correlation or 1.0. As correlation increases (moving from one scenario to the other perfect correlation), the junior tranches will less expensive or a lower spread. The senior tranche will be more expensive or have a higer spread. The middle tranches will be ambigious where there will be no definitive result.

P[no default] is the probability of no defaults. This is (1-1stTracnhe)^10=(1-.01)^10=.99^10. This represents the total will not default.

Junior P[>1 default] is probability of one or more defaults among the 10 bonds. Or the prbabikity of triggering the first or junior tranche. Under iid or independence, it is 9.96%.

The perfect is easier is easier as they all default or survive. That is 99% or 1% will default.

So as correlation increases, the probability of first to default goes down. The junior tranches are less expensive and have a lower spread.

As you look at the senior tranche, the probability of triggering default on the 10th tranche.

Senior P[all default] = 10th tranche ^ 10 = 1%^10=0% or all tranches to default

This is also calculated where 1%*1%*1%*1%*1%*1%*1%*1%*1%*1%=1%^10

The chances for perfect is 1%. As correlation increases from 0 to 1%, the senior tranche becomes more expensive. it has a higher spread because the probability of default increases.

Refer to 6:10 of the video for a simple diagram of this relationship.

http://www.youtube.com/watch?v=LaOuSgeXBHE

 

 

http://www.linkedin.com/groups/First-Default-CDS-4249235.S.94876147

 

First to Default CDS

On this topic, it states that as the correlation of the underlying references in the baskets increases, the value decreases. This seems counterintuitive to me. My thinking, if the probability of default increases due to the increase in correlation, wouldn't the value of the CDS increase as well? Are they referring to the value of the buyer or seller?

Who can help?

 

3 comments

 

Jose Fernando

Jose Fernando M. • Hi Chadderdon
(I apologize for my English, it is not my native language).
The statement is correct: as the correlation of the underlying references in the baskets increases, the value of the 1st-to-default CDS decreases.
Why? Suppose the following two Basket CDS:
A- The basket has 10 references and the correlation among them is 0.
B- The basket has 10 references and the correlation among them is 1.
If both A and B references have the same PD, which CDS is cheaper?
As the the references of the 'B basket CDS' have correlation 1, it is as a CDS with a unique reference. So, it is more likely to have a default among the references of A (there are more chances) than the references of B. That's why B is cheaper.
You will find the same behavior within CDO tranches!
It is not easy to explain, so try to think about it with constant PD and it will be easy to understand.
Hope I could help.

 

Michael

Michael O. • The way I like to think of it is if the correlations are perfect, the chance of the first one defaulting is the chance of any particular one, on their own defaulting. If they are completely independent, the chances of one defaulting =1-prob of NONE defaulting. The prob of none defaulting would be extremely low if they were all completely independent.

For exmaple, say there are 2 CDSs with perfect correlation and they each, individually, have a 10% chance of defaulting. This means that the prob that either one (the first one) defaults is exactly 10%. If they are independent, that means the chance of neither of them defaulting is 0.9*0.9=0.81. This means the chance of one of them defaulting has to be 19%. Therefore, it will possibly be worth more (to the buyer) because there is a greater chance of default (he gets paid out). This also means that the seller can charge more for it, which makes it more valuable to him. I think that is what they mean by the "value" being greater.

That is my 2 cents, anyway.

 

 

===============================

http://www.investopedia.com/terms/c/correlation.asp#axzz26PIZ7rQR

Definition of 'Correlation'

In the world of finance, a statistical measure of how two securities move in relation to each other. Correlations are used in advanced portfolio management.

Investopedia explains 'Correlation'

Correlation is computed into what is known as the correlation coefficient, which ranges between -1 and +1. Perfect positive correlation (a correlation co-efficient of +1) implies that as one security moves, either up or down, the other security will move in lockstep, in the same direction. Alternatively, perfect negative correlation means that if one security moves in either direction the security that is perfectly negatively correlated will move in the opposite direction. If the correlation is 0, the movements of the securities are said to have no correlation; they are completely random.

In real life, perfectly correlated securities are rare, rather you will find securities with some degree of correlation.