Exam Feedback October 2020 Part 2 Exam Feedback

[chintanmatalia]

Also for regression hedging I am confused between answer 137 or 187/9 ish( I chose second one) . Overall, questions that I seem to have messed up are default correlation, regression hedge, security selection, mistake in choosing answer carry trade (it should be riding the yield curve), CVA calculation, hedge fund question where I chose top down approach, marginal cost of funds answer, collateral, contingency fund planning, sharpe ratio,mortgage,EVT,RAROC(Doubtful),vasicek. Also would surely have messed up in some long para questions.

 

[matt.sutula]

That was a doozy for sure! Completely agree with everyone here, especially the fact that the GARP practice exam only had about 1-3 questions with overlap - what a crock! I figure I nailed about 1/3 of the questions, made educated guesses (narrowing down to 2 options) on the other 1/3 and completely winged the last 1/3 (with about 5 wild guesses).

I was very confident and ready to take this, but the test definitely threw a curveball with the complexity and difficulty of the questions. Took hundreds of practice questions from BT which obviously felt more in line with test difficulty. I supplemented my study with Kaplan (free materials as work paid for these), which were god-awful and not sure how those practice tests and questions would prepare anyone for this test - good thing i stopped using those materials the last month of study.

Lastly, echo same comments on hoping that curve gets moved down substantially. Good luck to all and fingers crossed for you all on 11/29.

 

[ngocle314]

Just wonder, provide the swchser Qbank ‘s poor quality, does swcheser notes cover enough the theory in the garp books ? Im rely on swchser note so much

 

[ps_ricky_son]

agree with any one of you that the exam is tricky and tough, especially for those mini-case studies which we were given a scenario and being asked which is correct. It seems that the choice of options represent different concepts, probably not from the same topic. I am just curious to know how the questions are counted to satisfy the weighting allocation, i.e. market risk 20%, credit risk 20%, etc. If I do not understand wrong, that weighting percentage means to test the relevant topic but not cross topics. Otherwise, the guidance of weighting makes no use as we could be tested different topics. Please correct me if I interpret it wrong. For e.g., one question is about the change in relation to the policy change of car loans in Thailand. We were asked the concept of systematic risk, hedging, (market risk topic) , asset/liability duration and balance sheet management which I do not recall exactly (liquidity risk topic?)...really many many cross topics...In fact, such questions suit the real world situation more.. however, should the GARP better hinted us beforehand in the 2020 mock exam or at least update a few questions to reflect not only the curriculum change but also the exam style change? If you have practiced several years of mock exams, they are just copying the questions....They should also consider those non-native English participants as at least 5 questions are of similar short essay style. Overall, this is not a good experience if they simply want to show that something bas been made to draw attention of potential candidates.

the other thing I wanna say is about the content being tested.. not asking about really important areas but sorts of relatively unimportant or even areas you may overlook. ..for e.g., one question on the most frequent type of operation risk for different businesses lines... the other relates to 72.5% of SA.....have to guess honestly..

time is indeed not a concern but the more time you could make it to double check your answers, the more frustration you make.

In view of the relatively fewer number of participants and the increasing difficulty of the exam, just curious how GARP will set the passing score... wondering if 60% - 70% still a good cut off?

and lastly, just received an email from the GARP asking us to provide some feedback on the exam.. not yet have a look and really no idea of the purpose... to arrange November exam better?

 

[VWJETTY]

In Libor question, the first choice said that Libor is based on actual transactions and the other half statement was true. But if you read schweser notes it says that Libor is not based on actual transactions and is one of the disadvantage as it is easy to manipulate

EXACTLY! But I read the other ones and none of them made any sense to me at all. So I just picked A and moved on. For me, at this question was towards the end and I had skipped the budgeting and Merton Model question so I didn't have time to sit there and meditate an answer haha!

 

[chintanmatalia]

Does anyone remember the answer for amount of collateral to be posted?

 

[chintanmatalia]

EXACTLY! But I read the other ones and none of them made any sense to me at all. So I just picked A and moved on. For me, at this question was towards the end and I had skipped the budgeting and Merton Model question so I didn't have time to sit there and meditate an answer haha!

Oh okay!

 

[chintanmatalia]

Does anyone remember the answer for amount of collateral to be posted?

I remember selecting 0

 

[matt.sutula]

Just wonder, provide the swchser Qbank ‘s poor quality, does swcheser notes cover enough the theory in the garp books ? Im rely on swchser note so much

I thought the notes were good as I used those initially and in conjunction with the BT materials. If I wasn't understanding a specific item/formula via BT, I would reference the Kaplan write-up, which seemed to help for a few concepts. However, the depth and detail that BT provides is spot on once you understand the concept. Some of the formulas and practice calculation questions Kaplan had in their q-bank weren't applicable or in line with the LOCs, or their questions seemed too basic. If I had to take this again, I would probably solely rely on BT materials/videos/etc.

 

[chintanmatalia]

Had the same issue with the Merton model, i.e. the result I obtained based on the correct formula did not match one of the answers. So I ended up choosing 21mn (which would be the result when not discounting the debt), however that is likely wrong.

Yeah even I chose 21

 

[nrustamli]

One of my colleagues who took the test before already warned me about the complicated questions, so I was not surprised. But still, it was incredibly and unnecessarily convoluted. So much so that I could barely remember 5 or 6 questions after the exam ended.
Regarding the Merton model question - I tested two ways, first I discounted 40 to PV and multiplied with N(d) value and subtracted it from 60 multiplied by N(d), no answer matched that, then I just used 40*57.5=23 and multiplied it by N(d) and subtracted from 60*N(d), which I believe returned one of the options.

one other question that drove me crazy was that 70+30m portfolio, but in the end I guess we should have just ignored return values and assumed it zero, then it would be B I think?

 

[SheOlya]

Do you happen to recall the factor values?
Think we just plug into r0+k(theta-r)DT+sigma*dw
I can't actually remember.

I still remember this question, because I felt very confused. The question was what is the expected 3 month short rate 3 years from now. The 3 month theta was 3.5%, current r was 1.5% and k=0.04. One of the answers was 1.73%, another 1.5..%, then 3.27% and something like 3.4...%. Does anyone know what is the correct answer?

 

[SheOlya]

I remember selecting 0

I remember choosing non zero because minimum transfer amount and rounding were 0.

 

[matt.sutula]

I remember choosing non zero because minimum transfer amount and threshold were 0.

Me too. I believe I chose 300k (maybe B) since that was the difference in changes between the increase in the MTM/position vs. the collateral.

 

[goodbunny]

I remember choosing non zero because minimum transfer amount and threshold were 0.

Yeah I am sure the difference was greater than MTA, so collateral had to be posted no?

 

[VWJETTY]

I remember choosing non zero because minimum transfer amount and threshold were 0.

Yeah I chose a non-zero as well. Because It breached the threshold (un-collateraized amount) so you need to fill it back up to the threshold amount.

 

[SheOlya]

With qq-plot, I forgot was it light tails or haivy tails?

 

[clement]

I also chose non-zero because MTA and rounding were both as "none". Did I have a different exam than @SheOlya? I think my answer was 700k

 

[SheOlya]

I remember only some questions:

1. QQ-plot
2. liquidity adjusted duration
3. Asset/security allocation
4. Risk contribution
5. Choosing one portfolio (I chose the one with highest Sharpe ratio, expected return was for all 4- 0.9)
6. 2 questions related to one story about CCP
7. Calculate correlation based on joint PD and 2 separated PDs.
8. S(t)-S(t-1)=a(mu-S(t-1)) calculate S(t)
9. Cross-currency swap
10. Incremental VaR
11. Var after stressed event with multiplier 3, SVaR and VaR for different confidence intervals were given
12. RAROC
13. Adjusted RAROC (accept/reject)
14. Effective ERM based on the scenario described
15. VaR of portfolio
16. Compare portfolio to benchmark(hazard rate, CET was given while for portfolio Capital and RWA were given), such as PD, CET ratio, and 2 other parameters (I picked one of 2 others to be the answer)
17. Cyber-security
18. Drawback of LIBOR
19. Merton model
20. Ho - Lee model
21. Vasicek model
22. DV01 Hedge
23. Climate risk
24. ML/AI question
25. Marginal VaR
26. Posted collateral
27. CVA calculation
28. Surplus question
29. Marginal Cost Approach calculation
30. Something in lines of specific internal fraud in which part of the bank it relates, like Corporate Finance, Treasury if I remeber correctly

Many qualitative questions that after reading a long story with useless info you read final question and try to answer based on the question.

 

[chintanmatalia]

With qq-plot, I forgot was it light tails or haivy tails?

Both of the tails were heavy