Exam Feedback January 2021 Part 2 Exam Feedback (postponed 2020 exam)

mattheweidinger

Active Member
Not sure if you're around these parts anymore since you passed (yay!) but I was hoping to calibrate my expectations ahead of sitting next week - would you say the exam difficulty is comparable to BT practice exams (which are supposed to match the test), the BT PQs (supposed to be harder), or even harder than BT PQs?
Happy top help. I'll PM you
 

mattheweidinger

Active Member
Ok. Due to the number of people who have messaged me and VOcker's request above, here we go:

Hey,

In terms of the exam...as I said in the beginning of the feedback thread:

I found part 2 very tough. The first 25-30 questions were so unordinary that I swear I checked the front of my booklet to make sure they had given me the FRM part II and not ERP part II test. After the first 30 or so questions were done, the back half of the exam was much more manageable although there were some tough questions placed there too. Looking back, it would have made more sense to have done the entire test backwards from 80 to 1 instead of front cover to back cover. ;) . I finished with 15 minutes left to review and quickly do the questions I skipped.

I managed to finish all the questions and I can say there were 20-26 that I had issues with - most of which were in the heavy front loaded part of the test. Many of the the toughest questions combined information from multiple volumes of the curriculum in the question stem and then asked a question prompt like "Which one of the following is correct/incorrect?" without providing any guidance regarding what to look for. Answering correctly required knowledge from all topics/volumes referenced in the question stem to either prove or disprove one of the four answer choices. (See below)

For instance, one question had Market risk information (Backtesting VaR), Operational Risk information (Basel capital requirements, Var & SVaR) and discussion of illiquid asset holdings in some investment fund. The question prompt here required knowledge of all three subject areas in order to either prove or disprove the answer choices listed below. Nothing like that existed on any of the GARP provided tests.

BT material was without a doubt the best prep for the part 2 test in my opinion, since most of the questions involved twisting your mind backwards, upside down and then dropping an anvil onto it. Looking back at my textbooks the day after the test, many of the questions relied on in depth knowledge coming from 1-2 sentences of the relevant section of the GARP textbooks.

There are many questions that are singular (4- 5 line paragraph) and just straight up tough because they rely on correctly remembering 1-2 sentences taken directly from the GARP official textbooks that don’t necessarily stand out as relating to the main point of the reading(s) in question. I found that many of these “easy” to “medium” type questions had 2 clearly correct answers.

----------------------------

For the toughest questions, a big Item set - "CFA L2/L3 style" is provided. But, it contains information from multiple volumes of the curriculum bundled in an interconnected manner in there (Market, Credit, Liquidity and Investment for example). Here's an example that has stuck with me (names changed for convenience, but overall structure is basically the same...):

Hedge Fund XYZ has a long position in shares of ABC Inc, a large global manufacturing firm. The position’s alpha is currently 5.5%, Sharpe Ratio is 0.7, Beta is 1.6, MCaR is 0.3, risk aversion is 0.9 and active risk is 20%. The fund is a distressed debt fund.

The CIO and CEO are friends from business school and they often outsource much of their trading, reporting and custodial services to a third friend at various firms he runs including a brokerage, a bank and a trust service. The friend is also from the same business school as the CIO and CEO. The fund has existed for 10 years and has never reported less than 30% after tax, after fee annual returns to clients. The average Distressed fund reports 10%.

An analyst at the hedge fund is analyzing his recent trade on Bank Trump and needs to determine if his analysis of the bank is accurate. He is given the bank’s VaR and SVaR for the most recent day as well as the 60 days average for both VaR and SVaR. The last time the analyst conducted a backtest of the banks 95% VaR at a 99% confidence interval, he noted that there were 15 exceptions based on 500 trading days. Bank Trump holds 1.2B in total risk capital.

*Table displayed below with a bunch of the stats referred to above and some random, totally IRRELEVANT figures to throw you off*

Bank Trump is located in Kazakhstan, which recently experienced a military coup that deposed the previous government. The analyst is considering investing in a bank in Zimbabwe or one in Ghana as well.

After completing his analysis, the analyst is invited into the CIO’s office to discuss an investment idea. The CIO describes the bank in Zimbabwe that the analyst is considering taking a position in. They discuss how the bank has recently seen its share price decrease substantially as a result of losses in its corporate loan portfolio.

Question prompt then asks: "Which one of the following is not incorrect?"

A) The hedge fund should sell it’s position in ABC inc, if the alpha increases another 5%, but maintain it, if the alpha rises only 4.5% more.

B) The bank in Kazakhstan has a higher investment risk than Ghana according to the 2010 United Nations rule of law Index, but lower risk than the bank in Zimbabwe according to the same index.

C) The bank’s capital requirements are in compliance with Basel I, but are not in compliance with Basel II, Basel II.5 or Basel III.

D) The hedge fund likely has drawn the attention of the FBI financial crimes division.

-----------------


This is a shortened version. In reality, this would cover 80-90% of the page. I would say 15-20% of questions were like this. Some with one question asked at the end, others with 2. You essentially need to know investment risk to prove or disprove choice A; country credit risk to prove or disprove choice B; market risk AND operational risk (All versions of Basel capital requirements) to prove or disprove choice C; and another section of investment risk to prove or disprove choice D.

There was another one like this in the first 5-10 questions of the exam which asked you to do a ton of stuff related to MVaR, RAROC, Adjusted-RAROC. The text and tables took up the entire left page and left some space on the right page of the booklet for two questions to be asked.

In the end, you need to do all of the BT practice tests and questions to prep for this. I did all the GARP questions and they are totally crap. NOTHING like the real thing - this was mostly true on FRM 1 when I wrote in November 2019 as well. However, the part 2 was TOTALLY different in terms of the question style and difficulty level. I don't even think BT fully captures the difficulty of the actual test, since most BT questions are stand alone and do not mix in material from across different volumes in the curriculum into the same question. Still, they are the closest thing to the actual test day difficulty level you will come across.

Maybe this is a recent change GARP made to increase the difficulty of the tests, so maybe study materials will adapt in the coming years...

Anyways. Let me know if you have any specific questions. I got my pass results in Feb and am expected the mailed certificate sometime soon.
 

LucreziaB

Member
Thanks! Yeah, that's a doozy but super helpful to have an idea of in advance. Once you are able to break out its components, BT questions seem to be good preparation for each of the pieces.

All that said, what do they really think they're accomplishing by making such a question multiple-choice?! (I could see it being a good essay question where the perfect score would show perfect mastery of all the concepts but there would be partial credit for mastering most of them.) You have to know several different parts of the syllabus just to narrow such a question down to two answers at which point your 50% chance of getting it right is marginally better than someone who knows absolutely nothing and has a 25% chance of randomly selecting the right bubble. (whereas the same person would score very low on an essay response). Frustrating for an 80-question test, flies in the face of all the readings on validating rating models, backtesting, etc. :)
 

mattheweidinger

Active Member
Thanks! Yeah, that's a doozy but super helpful to have an idea of in advance. Once you are able to break out its components, BT questions seem to be good preparation for each of the pieces.

All that said, what do they really think they're accomplishing by making such a question multiple-choice?! (I could see it being a good essay question where the perfect score would show perfect mastery of all the concepts but there would be partial credit for mastering most of them.) You have to know several different parts of the syllabus just to narrow such a question down to two answers at which point your 50% chance of getting it right is marginally better than someone who knows absolutely nothing and has a 25% chance of randomly selecting the right bubble. (whereas the same person would score very low on an essay response). Frustrating for an 80-question test, flies in the face of all the readings on validating rating models, backtesting, etc. :)



I think that the tests started to become more difficult in recent years. Especially now with more GARP registrants and more topics to cover in the curriculum, 80 questions for 80 topics or learning objectives (more or less) does not permit them to appropriately test all the material. But, if they test 2-4 different topic areas (one per question answer choice...) X 80 questions = 160-240 testable topics essentially.
 
Top