Exam Feedback November 2018 Part 2 Exam Feedback

max8126

New Member
This is a part of the waterfall structure. The OC would first protect the Equity tranche up until a breaking point where the losses are more than the overcollaterisation and start eating up to the waterfall structure. I answered waterfall structure as the question clearly mentioned protect the class A throughout the lifespan of the securitisation (Pool insurance is true but does not protect throughout the lifespan - only kicks in if the level of defaults are very2 high etc; what if default are not that high)
Waterfall is simply the order of payment and loss absorption, like someone mentioned before it's just a feature of securitization.
The credit risk book has several sections on OC. It should be clear that it's there to protect senior tranche, often at the cost of the equity tranche.
 

terrance00232

New Member
Hi guys,
I remember there's a question which need to calculate the four steps Binomial Tree. (semiannually payment in 2 years !?)
Is there anyone can recall that !?

this examination is still so hard for me,
but I'm glad to know that someone who think it's easy
 

rana.nadeem

New Member
Hi guys,
I remember there's a question which need to calculate the four steps Binomial Tree. (semiannually payment in 2 years !?)
Is there anyone can recall that !?

this examination is still so hard for me,
but I'm glad to know that someone who think it's easy
It was 1 year spot rate at 3% initially, and then 1 year forward rates (1 year from now) are given as 3.8% (up node) and 2.8% (down node). The compounding is to be done semi annually. Equal real world probabilities are to be assumed to get present value of 2 year ZCB.
 

firsova

New Member
There were many questions related to CCP. Does anybody remember those? Also, there was question for which i chose Clustering as the answer... 2/3 questions on role of dollar, question on Merton, KMV model, question on friction from securitization...

I remember the question where I answered clustering as well
 

matt6558

Member
I think there’s a question on iasb/fasb treatment in difference on ecl/theres a question on spread question given the default rate/fv and mv value, rf rate.
 

rahulkp28

New Member
It was 1 year spot rate at 3% initially, and then 1 year forward rates (1 year from now) are given as 3.8% (up node) and 2.8% (down node). The compounding is to be done semi annually. Equal real world probabilities are to be assumed to get present value of 2 year ZCB.
I remember the question where I answered clustering as well

This question pertains to which technique would bank's use to segregate borrower characteristics such as credit history, score etc.? Options were clustering, K fold, penalized regression, classification. I chose Clustering.
 

haziqnazri

New Member
there was one question on the data of banking applications.... what was the best machine learning tools to use to extract the data? I answered clustering
 

highwayone

New Member
there was one question on the data of banking applications.... what was the best machine learning tools to use to extract the data? I answered clustering
The answer can't be clustering. The question is to predict PD, which is a y label. You should choose a regression model.
 

highwayone

New Member
Waterfall is simply the order of payment and loss absorption, like someone mentioned before it's just a feature of securitization.
The credit risk book has several sections on OC. It should be clear that it's there to protect senior tranche, often at the cost of the equity tranche.

OC is used to protect equity tranche. For example, the par value of the backed asserts are 100 million, and a MBS is issued with a value of 99 million. Then 1 million can be lost before any damage to the equity tranche.

Waterfall is used to protect senior tranche. Because the equity tranche absorb the loss first then the mezzanine tranche.
 

firsova

New Member
Waterfall is simply the order of payment and loss absorption, like someone mentioned before it's just a feature of securitization.
The credit risk book has several sections on OC. It should be clear that it's there to protect senior tranche, often at the cost of the equity tranche.

I know that o/c is an enhancement, however I recall that the question was structured in a way that made me doubtful about o/c. This is what Malz says:
1542695586850.png
 

max8126

New Member
OC is used to protect equity tranche. For example, the par value of the backed asserts are 100 million, and a MBS is issued with a value of 99 million. Then 1 million can be lost before any damage to the equity tranche.

Waterfall is used to protect senior tranche. Because the equity tranche absorb the loss first then the mezzanine tranche.
I believe while your example captures the spirit of OC it's oversimplified. OC test is done on tranche level. As that 1 mil is being chewed through most likely the senior tranche would breach the OC threshold first (since it has the smallest denominator) and trigger OC protection. Only at maturity is unused OC released to equity. So in a way yes equity is also protected, but not before senior tranche.

Another angle to think about this is that the need to OC in reality is because to earn certain rating some minimal OC ratio is required. Why? Because OC protects senior tranche. The equity tranche is there precisely as a high risk high return option to investors. Wouldn't make sense to have some form of internal credit enhancement that's there to protect equity tranche first/only.
 

runal

New Member
There another question on the 95% var for basel 2.5 where there are 20 exceptions for 250-days VaR, i really thought there's 2 answers B and C where the C) Reject the VaR model as the exception exceed the significance level at 95% confidence interval and also B) the basel 2.5 Var was higher than 9 million ish.....

As far as i remember the var came around 7 million and i marked C) C) Reject the VaR model...
 

highwayone

New Member
If I recall well, the question was to find out the best approach to apply when you have data and you do not know its characteristics

Clustering identifies outliers that do not have strong connections with the rest of the data. You can't predict PD with clustering only.

For example , you can use clustering to divide the applicants into 10 classes. But without any PD information about some applicants in each class. You will not know the PD of each class.
 
Top