Exam Feedback November 2021 Part 2 Exam Feedback

[Nicole Seaman]

This is a reminder that we cannot allow the discussion of the Part 2 exams in the forum when they are still ongoing. The Part 2 exams continue through the end of this week.

We will open this exam feedback thread for comments and discussion when the exams are over. Please do not post any exam feedback in other areas of the forum, as we need to moderate all of the posts and delete them if they pertain to the current Part 2 exams.

 

[Nicole Seaman]

This thread is now open for discussion! We hope that everyone did well on the FRM Part 2 exam! We would love to hear any feedback that you have about the exam, especially any helpful information that you can provide about the new CBT format. How did the exam go? Did you encounter unexpected questions? How was the experience with the COVID guidelines? Thank you in advance for any feedback you can provide!

 

[nc27]

Overall I think GARP did a good job to cover a lot of LOs for my exam (30% quant / 70% qualitative)... as many have said before, for a lot of qualitative questions 2 answers seemed ok, so that was a bit frustrating. The CBT format is alright but no score was shown at the end of the exam. So now, I hope for the best! I will take some rest, this exam has taken its toll on me.

EDIT 1: BT practice bank, I must say, is very well aligned on what is being asked on exam day... so as usual, good job BT team! you deserve your fame and first place in the realm of prep provider for the FRM exam

EDIT 2: Questions were not tedious to read, in fact they were pretty concise, but possible answers were very tricky

EDIT 3: for future exam takers, do not underestimate this exam, it is brutal even if it is qualitative in nature

 

[Chacko_aspire]

First off, I didn't have any software or internet issues (being in India, that is a valid concern and the electricity did go off in the building towards the end of the exam). Invigilation style of the exam was a new experience. The exam supervisor (my name for him) said that I am being recorded from an overhead camera and that that recording will be checked (by someone). I am not sure that this was to be expected, or if this is legit going to happen but I didn't press him further on the topic.
The CBT format was good in terms of ease of use, especially in terms of flagging and reviewing questions. But, I couldn't skim through the question paper to get an overall feel of the difficulty level, as I could have done with paper based exam.
In terms of the exam contents, @nc27 seems to have covered my observations as well. Additionally, some of the questions seemed to be from obscure portions that weren't tested previously as well. And one question was a combination of previously tested concepts of different LOs, with a question statement that was vague on what was actually being tested.
The BT practice and mock exams helped in preparing me for the quantitative concepts very well. The qualitative questions were quite a challenge, still. To memorise and recall so many different of fields of expertise is a struggle.
Hope that my efforts bear fruit!

@David Harper CFA FRM Appreciate the effort that your team and you puts in delivering quality material on such topics; and the breadth, depth and quality of discussion on each LO practice paper in various corners of the forum helped to answer questions that I myself didn't know I must ask.

 

[kawal_frm]

35% quant...65% qualitative.

Couple of quant questions gave more info than required(probably to confuse). Quant questions were much easier for someone who has practiced BT.

at least 10 credit risk questions were quant. 50% MR quant. OR heavily qualitative.

Some of the qualitative questions, they mixed different LOS as 4 options and you have to pick the correct one.

Was able to do 50 out of 80 in the first go. Hoping strike rate of at least 80% here...so 40 correct

For another 20 I was able to filter out 2 out of 4 options. Hoping at least 8 correct here.

for the rest 10, it was more like a wild/educated guess(most of these questions were mainly from theoretical chapters of operational risk which was indeed my weakest part). Hoping at least 2 out of 10 are correct here.

So all in all hoping to get at least 50 correct. If some of the Wild/educated guesses goes my way, then might touch 60.

 

[ambrosianas]

Is it just me or did anyone else feel there was no correct answer for the lognormal VaR question? I thought it was a super easy question but after calculating it twice none of the choices came close.

 

[ambrosianas]

While I think the materials was very quant in nature especially in credit risk, market risk and Basel Accord, the test was highly qualitative and ambiguous… I was very well prepared for complex multiple step calcs but I felt 90% of the test was qualitative and difficult to choose between two answers.

I don't think the exam covered the materials well. In fact, a lot quant oriented materials were absent. Almost nothing on Basel Accord. I walked out of the room feeling like I just took an exam about best practices in corporate culture. In my opinion, this was an "ops risk/current issues" heavy test, not really fair for those who studied market/credit quant models well.

 

[hateeque]

Does 50 give you a pass? I think not

 

[FabbrisF]

Here are some questions I remember and the answers I gave:
1) Model 1, what is the volatility of the rate in the tree? I answered sigma*sqrt(1/12) = 0.24%
2) BCVA change (tricky) answered -53,000 (B)
3) Model 3, answered "non parallel shifts"
4) 3 year survival prob given hazard, answer: exp (-hazard*3)
5) structured product waterfall, credit enhancement senior 15K and equity 733K
6) Copula, maps into normal distribution (answer A)
7) Merton probability of Default using forumla PD= N(...), got 38.8% answer D
8) Green Swan, is difficult to predict and consequences are difficult to model (answer c)
9) FX options, imply more observation in both tails than lognormal
10) Reduce counterparty risk --> I answered "risk mutualization"
11) LIBOR/SOFR --> SOFR reflects changes in collateral markets
12) Basel Credit and Market Risk --> Market capital charge 23.7m answer B
13) CVaR, had to calculate beta with given correlation
14) UL of a portfolio given UL1 and UL2 and correl
15) CVA increases as recovery rate decreases
16) active risk aversion given information ratio and tracking error, answered 0.1
17) Surplus at risk, calculate vol of surplus cant remember the answer

Feel free to challenge answers and add more for completeness.

 

[Hamam]

Does 50 give you a pass? I think not

Here are some questions I remember and the answers I gave:
1) Model 1, what is the volatility of the rate in the tree? I answered sigma*sqrt(1/12) = 0.24%
2) BCVA change (tricky) answered -53,000 (B)
3) Model 3, answered "non parallel shifts"
4) 3 year survival prob given hazard, answer: exp (-hazard*3)
5) structured product waterfall, credit enhancement senior 15K and equity 733K
6) Copula, maps into normal distribution (answer A)
7) Merton probability of Default using forumla PD= N(...), got 38.8% answer D
8) Green Swan, is difficult to predict and consequences are difficult to model (answer c)
9) FX options, imply more observation in both tails than lognormal
10) Reduce counterparty risk --> I answered "risk mutualization"
11) LIBOR/SOFR --> SOFR reflects changes in collateral markets
12) Basel Credit and Market Risk --> Market capital charge 23.7m answer B
13) CVaR, had to calculate beta with given correlation
14) UL of a portfolio given UL1 and UL2 and correl
15) CVA increases as recovery rate decreases
16) active risk aversion given information ratio and tracking error, answered 0.1
17) Surplus at risk, calculate vol of surplus cant remember the answer

Feel free to challenge answers and add more for completeness.

I think there were two different exams because some of the questions here weren’t asked on my exam

 

[Frodo81]

Apologies for having shared my impressions already before the exams have been finished (@Nicole Seaman) - my bad, was not really aware of it. However, I will quickly summarize again now.

Overall, the exam was pretty much as expected after intensive months of preparation. A lot of qualitative questions (would guess around 70%) and covering topics I knew (or I knew that I should have known... ). But, not really a surprise to me either, the answers weren't exactly obvious. The quantitative questions were mostly straight forward.

Compared to the level I the 4 hours time was much less of an issue this time which helped me a lot to be more relaxed and focused overall. Also, the CBT format is, in my view, more convenient than the paper based exams. On our site we could start individually (i.e. whenever we were ready), you can flag questions and easily return to them at the end.

At the end of the day I won't be on the podium but I did what I could in terms of preparation and compared to the level I I am slightly more relaxed. And so I agree to the (unknown) girl that said to me right after the exam: Hope to never see you again...

 

[kawal_frm]

Here are some questions I remember and the answers I gave:
1) Model 1, what is the volatility of the rate in the tree? I answered sigma*sqrt(1/12) = 0.24%
2) BCVA change (tricky) answered -53,000 (B)
3) Model 3, answered "non parallel shifts"
4) 3 year survival prob given hazard, answer: exp (-hazard*3)
5) structured product waterfall, credit enhancement senior 15K and equity 733K
6) Copula, maps into normal distribution (answer A)
7) Merton probability of Default using forumla PD= N(...), got 38.8% answer D
8) Green Swan, is difficult to predict and consequences are difficult to model (answer c)
9) FX options, imply more observation in both tails than lognormal
10) Reduce counterparty risk --> I answered "risk mutualization"
11) LIBOR/SOFR --> SOFR reflects changes in collateral markets
12) Basel Credit and Market Risk --> Market capital charge 23.7m answer B
13) CVaR, had to calculate beta with given correlation
14) UL of a portfolio given UL1 and UL2 and correl
15) CVA increases as recovery rate decreases
16) active risk aversion given information ratio and tracking error, answered 0.1
17) Surplus at risk, calculate vol of surplus cant remember the answer

Feel free to challenge answers and add more for completeness.

I got completely different exam. No BCVA. No surplus as risk. No question on structured waterfall. Nothing on risk aversion. I had question on cox ingersol. Nothing on model 1 or model 3.

I did get Basel market risk 23.7 option B. And 3 year survival probability with given hazard = 8%.

 

[nc27]

I got completely different exam. No BCVA. No surplus as risk. No question on structured waterfall. Nothing on risk aversion. I had question on cox ingersol. Nothing on model 1 or model 3.

Yes it seems that we had the same exam @kawal_frm!

 

[kawal_frm]

Does 50 give you a pass? I think not

I think cut off should be 55+. Exam wasn’t difficult. It was very conceptual. For most question you could have easily filtered out 2 incorrect options.

 

[nc27]

I think cut off should be 55+. Exam wasn’t difficult. It was very conceptual. For most question you could have easily filtered out 2 incorrect options.

For part 1 I thought the cutoff to be low 60 but then it comes to be low 50.. So who knows, I personally think that low 50 will be the cutoff for part 2

 

[kawal_frm]

Yes it seems that we had the same exam @kawal_frm!

Distance to default between x y and z.
2 questions on Merton debt and equity calculation.
1 question on hurdle rate. 1 on raroc.
1 question on LDA. 2.66 answer
2 year bond credit spread question. 1.69 answer
Threshold. I answered 0
Ul with .25 correlation.
2 million cny. 120 loans. 7 defaults at 95%. Find loss at 95%
2m loan. Loss I answered as less than 2m based on LGD. Option D

 

[kawal_frm]

For part 1 I thought the cutoff to be low 60 but then it comes to be low 50.. So who knows, I personally think that low 50 will be the cutoff for part 2

For 25 numerical questions ; they were even easier than Garp mock exam and anyone with even basic knowledge should get atleast 20 correct.

For rest 55 almost in every question 2 of the options were complete non sense and you are down to picking 1 out of rest 2. Even random guess will get you 25 correct. So someone who spent 150-200 hours should get 45 right easily. I am fairly certain that only 50 correct is fail.

 

[nc27]

For part 1 I thought the cutoff to be low 60 but then it comes to be los 50.. So who knows I personally think that low 50 will be the cutoff

For 25 numerical questions ; they were even easier than Garp mock exam and anyone with even basic knowledge should get atleast 20 correct.

For rest 55 almost in every question 2 of the options were complete non sense and you are down to picking 1 out of rest 2. Even random guess will get you 25 correct. So someone who spent 150-200 hours should get 45 right easily. I am fairly certain that only 50 correct is fail.

Yes, I thought that too for part 1 but not as many people as we think prep with BT so we might be skewed in our judgment

 

[nc27]

Distance to default between x y and z.
2 questions on Merton debt and equity calculation.
1 question on hurdle rate. 1 on raroc.
1 question on LDA. 2.66 answer
2 year bond credit spread question. 1.69 answer
Threshold. I answered 0
Ul with .25 correlation.
2 million cny. 120 loans. 7 defaults at 95%. Find loss at 95%
2m loan. Loss I answered as less than 2m based on LGD. Option D

Yes same exam! The ones on raroc and cny loans left me clueless and a bit angry as I have worked on that a lot

 

[kawal_frm]

Yes same exam! The ones on raroc and cny loans left me clueless and a bit angry as I have worked on that a lot

I answered aRaroc reject. Hurdle rate 6.5. For CNY they already gave 7 defaults. So 7*2m is pd*ead at 95%. Just Multiply lgd.