Exam Feedback May 2016 Part 2 Exam Feedback

terzim

New Member
I found the exam extremely difficult. The mathematics involved in the quantitative questions was not trivial. The qualitative questions had often 2 answers that could both be largely correct (so the issue was identifying the MOST correct). The case studies as well... very long and painful.

I remember I got very upset on one OPERATIONAL RISK CAPITAL question... I thought that asking the Basic Indicator Approach (15%), but then splitting by business lines (similarly to the Standardized Approach), was just there to confuse people. And I am still unsure about the answer.

Other key highlights from my side:

=> many more CVA and counterparty risk questions than I expected (including direct CVA computation)
=> VaR was tested very heavily from all possible angles. The calculations were difficult especially when done on a calculator (I prepared my exam on Excel, and it is way easier to do calcs on a spreadsheet).
=> lots on Merton and KMV (structural models) but nothing or almost nothing on intensity models (I was expecting a CDS spread calculation and it took me ages to remember the formula, luckily it did not come).

That's all I can add compared to what was said before. Wish good luck to you all.
 

M.A.

New Member
Subscriber
PSA & the SMM question was about yhe the greates paid back could occure when... I do not recall the provided answeres but some was talking about the PSA 115 or PSA 90...
 

fujii_itsuki

New Member
I found the exam extremely difficult. The mathematics involved in the quantitative questions was not trivial. The qualitative questions had often 2 answers that could both be largely correct (so the issue was identifying the MOST correct). The case studies as well... very long and painful.

I remember I got very upset on one OPERATIONAL RISK CAPITAL question... I thought that asking the Basic Indicator Approach (15%), but then splitting by business lines (similarly to the Standardized Approach), was just there to confuse people. And I am still unsure about the answer.

Other key highlights from my side:

=> many more CVA and counterparty risk questions than I expected (including direct CVA computation)
=> VaR was tested very heavily from all possible angles. The calculations were difficult especially when done on a calculator (I prepared my exam on Excel, and it is way easier to do calcs on a spreadsheet).
=> lots on Merton and KMV (structural models) but nothing or almost nothing on intensity models (I was expecting a CDS spread calculation and it took me ages to remember the formula, luckily it did not come).

That's all I can add compared to what was said before. Wish good luck to you all.
I remembered for the basic indicator thing, I chose C, the one for sales and trading should be correct...

Hopeful I chose the right answer
 

sriniwas

New Member
PSA & the SMM question was about yhe the greates paid back could occure when... I do not recall the provided answeres but some was talking about the PSA 115 or PSA 90...
Answer for that worked out to be 115 PSA for me. Hope its right.
 

Ajay

New Member
There was a question on Lvar that tested for higher or lower under constant or exogenous approach.

For prepayment, 115 PSA appears to be correct. For PSA, prepayment increase by 0.2 % each month till it reaches 6%. After that is remains constant at 6 %. Since the question was for 115PSA, prepayment would be 1.15 times 6%. Forgot the formula for SMM. So hope 115PSA would generate the maximum prepayment amount the provided options.
 

jacksonxyz

New Member
Exam was definitely tough... Really scared of the results now!

I have used Schweser material for preparation. I have solved Schweser practice exams and GARP paper. Scored 50-65% in these exams (could have scored more, but was lazy to give these too seriously). I have started to practice BT chapter wise questions but gave up as I was running out of time for revision. Ideally, a little bit of more structured preparation with BT questions would have boosted my confidence in the exam.

Can remember questions on following topics:
1. Hurdle rate calculation using equity and preferred equity
2. RAROC
3. Netting factor calculation
4. QQ plot
5. CVA calculation
6. WCDR calculation
7. VaR - Incremental, Marginal VaR
8. Illiquidity
9. On PD and VaR on tranches in MBS
10. Mapping - Duration, cash flow, principal
11. Expected shortfall calculation
12. Correlation and PD
13. WWR and RWR
14. Pareto, negative binomial distribution, log normal
15. SMM, CPR, PSA
16. CCPs and default
17. Market risk with stressed VaR
18. LIBOR, OIS rates usage
19. Frechet distribution
20. Correlations and effect
21. IR, Vol - cal weight of Index position

Shall extend this list as I recollect more..

On QQ plot, I believe the answer is thin tails to the left and thick tails to the right. I have tried to compare the deviation of datapoints to get to this answer..
 
Last edited:

M.A.

New Member
Subscriber
What was the answer for the ES - 10 var s were provided - was it only sum and average by 10!!!

I remember a question regarding frechet and a shape more than 3 i answer it has heaver tails than Gumbel/ totally not sure. I remember that it is heaver than Std normal considering std is almost the same as Gumbel if the shape is 0
 
Last edited:

Ajay

New Member
What was the answer for the ES - 10 var s were provided - was it only sum and average by 10!!!

I remember a question regarding frechet and a shape more than 3 i answer it has heaver tails than Gumbel/ totally not sure. I remember that it is heaver than Std normal considering std is almost the same as Gumbel if the shape is 0
Are u referring to the question about ES - 10 vars were provided for 90 to 99%?

Shouldn't the answer be average of the last 5 vars? i,e average of vars for 95, 96,97,98,99% vars?
 

Revilo

New Member
Are u referring to the question about ES where 10 vars were provided for 90 to 99%. Shouldn't the answer be average of the last 5 vars? i,e average of vars for 95, 96,97,98,99% vars?

On my side i just abg 96-97-98-99 if i recall. I excluded 95level.
 

gaurav11235

New Member
Hi


Hi Gaurav,

18. I have also selected 50 million, but it was not working out exactly. There were 9 members with contributions of 10m each (90m). Than there was an initial margin of 10m. And a 10m first peice loss contribution by the CCP.
So if the total loss was 50 m, 10m from intitial margin, 10 million from contribution of the member who is defaulting (leaving 80m contributions from rest members), 10 million from first piece loss. So that leaves us with another 20 m loss to be set off.
If we deduct this 20 m from the 80 m contributions left over by the other non defaulted member, the balance should be 60m. But there wasn't a choice with 60m. I was frowning upon myself for not getting it right. Can anyone help.

Regarding current issues questions:
- There was one on which discount rate to use for derivative. I answeres t bill rates, because the derivatives had long maturity. other choices on repo, OIS and Libor were incorrect.
- There was also a question on lender of last resort vs liquidity regulation. Answer was C or D i guess.
There were actually 8 members only which explains answer as 50 m. There was a question on CCAR stress scenario where a portfolio had significant exposure to domestic securities. I answered A that risk manager should use the stress scenario developed by regulators.
 

gaurav11235

New Member
1. Does anyone remember Distance to Default problem?

S=20
X= 7+ 0.5x7 = 10.5
Volatility = 28%
R = 5%(???)
T = 1

So...DD = 2. 34

answer was " D. BB/ BBB- " (2.2 ~ 2.5)

right or wrong?


2. Does anyone remember Credit Var(number 2) problem?
I got the same answer.
 

M.A.

New Member
Subscriber
I think also there was a question which statement is true and a table is provided for Credi var and Market Var choices given were including a) credit var of 6m and b) Market stressed var equal 316M any one remember the answer
 

Revilo

New Member
For the 1st question, what did u guys reply? It was about the probability of default of a b c d with a correlation matrix. I chose the one with the highest positive correlation rAte. Which was 0.5..
 

M.A.

New Member
Subscriber
I found the exam extremely difficult. The mathematics involved in the quantitative questions was not trivial. The qualitative questions had often 2 answers that could both be largely correct (so the issue was identifying the MOST correct). The case studies as well... very long and painful.

I remember I got very upset on one OPERATIONAL RISK CAPITAL question... I thought that asking the Basic Indicator Approach (15%), but then splitting by business lines (similarly to the Standardized Approach), was just there to confuse people. And I am still unsure about the answer.

Other key highlights from my side:

=> many more CVA and counterparty risk questions than I expected (including direct CVA computation)
=> VaR was tested very heavily from all possible angles. The calculations were difficult especially when done on a calculator (I prepared my exam on Excel, and it is way easier to do calcs on a spreadsheet).
=> lots on Merton and KMV (structural models) but nothing or almost nothing on intensity models (I was expecting a CDS spread calculation and it took me ages to remember the formula, luckily it did not come).

That's all I can add compared to what was said before. Wish good luck to you all.
Isn't here you will be calculating them based on average of poitive 3 years x 15% but only on the business unit level or separately as the answere were given lead to doing it that way
 

Revilo

New Member
There were actually 8 members only which explains answer as 50 m. There was a question on CCAR stress scenario where a portfolio had significant exposure to domestic securities. I answered A that risk manager should use the stress scenario developed by regulators.
For ccar i went with The -2.
Isn't here you will be calculating them based on average of poitive 3 years x 15% but only on the business unit level or separately as the answere were given lead to doing it that way

Yeah i did that too. I replace the negative by 0 and ended up dividing by 2.
 
Top