Any Feedback on FRM 2009 Exam ?

[munish_bansal1]

@NewTurtle

It seems that either we studied in the same manner or we approach problems in a similar fashion. After reading your questions and answers, I am little more confident. The only difference is that I gave full FRM. Would like you to post few more if you can remember because I am not able to think anything other than what had already been discussed.

 

[David Harper CFA FRM]

@asja: thanks for your feedback. I am glad you were happy because, as you know, responding to your queries occupied a lot of my time. This is interesting: "GARP seems to most value how strongly the concepts are embedded in you so you know how to approach/think right away after reading the question.. and just memorizing all the formulas would not help much"

@Chinni. Thanks for your detailed feedback.
In regard to, "having an FRM might work against me," I have never heard that (just speaking for myself)...

@Jack, thanks for your thoughtful feedback. You have frankly summarized my frustration as an FRM course provider. Specifically, if the following is true: "really feel that a person who’s put in 500 hours of study could’ve done a worst job than a person who’s put in only 200 hours or study, and that’s what bothers me the most - it really wasn’t representative of what we know" ... then (in addition to the fact that I empathize, the exam should not engender this expericence for you!), it becomes challenging to focus the preparation (i.e., for both the candidate and the trainer like myself) ... I feel your comments speak to the partial sense of randomness vis-a-vis the exam ... I am grateful that you value the "learning for learning's sake," although (of course), the exam result should be correlated to the effort

@Anwar: Any idea how the GARP wil evaluate that questions as there were all answers wrong ?...I don't know (sorry)

@mdecav: "how has GARP handled the issue when a topic (i.e. the box spread question) is not explicit in the AIM?"
This irks me too to b/c box spread is in the Hull, but I did not include because box spread was noticeably omitted in the AIMs. This simply should not happen, IMO.
My understanding, based on one conversation last year with GARP about this, is: if a question is found to be flawed ex post (e.g., no answer is correct), then it is nullified. However, if the box spread question is correct but "merely" not formally assigned, my assumption is that it will still be retained.

@Kumar: In regard to: "1) You will be graded as PASS/FAIL, 2) You will not know your scores. So unlike last time, will there not be results in Quartiles??"
If you are asking me this as a question, then I do not know, apologies ... i do not recall any formal information given about the post-exam breakdown. (I guess i was thinking there might be a breakdown, but maybe i only imagined that, and perhaps, given the relatively low per-topic density--i.e., nine topics against 140 questions--it may be unrealistic?). Yes, of course the result is PASS/FAIL but beyond that, i do not know

@Dinesh,
Thank you for your kind feedback ... if you decide to come back and still need deferral assistance due to the emergency, don't hesitate to contact me (.(JavaScript must be enabled to view this email address)) and Suzanne (.(JavaScript must be enabled to view this email address)), we cooperate with emergencies

@Ajit,
Does GARP remove ambiguous questions (test of english/comprehension) and then grade the rest?
My assumption (i.e., I have no verified, formal understanding on this matter) is: if the question is ex post found to be a tangible error (i.e., no answer is correct), then it will be nullified. Otherwise, I assume it will NOT be nullified; e.g., I doubt very much they would nullify a question because the grammar is imperfect.
Re: do you suggest to start preparing for CFA Level 1 ? I am from technology/maths background: I don't have enough information to have an opinion (I am currently working on expanding the site to offer better career-related assistance, I hope we can be more helpful in this regard in the near future!)
Re: Your last week review notes were really helpful to me. Thanks a lot for all your support.
Thank you very much, I really appreciate. Although I wanted to give more help, but I just ran out of time.

@kkt: thanks for your input. I agree luck should not trump skill. (further, I would actually set the bar higher: I would hope the exam aspires to minimize the sheer advantage of an experienced practioner over a new student. Of course, experience will always help, but I believe the exam should faithfully test the ASSIGNMENTS so that a new learner, without deep risk experience, can know what to expect. Same idea as CFA, the CFA is not biased against new learners)

@benoit16, Thanks for your further input. About the transparency, I wholeheartedly agree, as a provider I totally support the value of transparency. But frankly, GARP will need to improve their own rigor before they can be transparent.
I must be dense, what happens on the 5th of January?

@munish: no, the desirability of option exerise (it is never optimal to early exercise an American call on non-dividend) does not imply put-call parity equivalence. There is an American put-call parity (Hull 9.4) but it remains different. I now see that Ashok adds, "[for the American option, it is] unlikely to be any dividend declared in the next 6 months," I am sorry, but that still does not reduce the American-style put-call parity to the European-style that we tend to memorize. Answer (d) confuses b/c there does exist a put-call-parity, so apologies, it confuses me....

@Patrick, Re: "I think it may be helpful for candidates if you can teach some skill of taking exam in the class on topics such as making educated guessing as the same time to cover all required knowlege."
That is good advice, I will seriously look into adding this component to our preparation.

@GW: your interpretation of this question makes the most sense to me; e.g., Hull p 211: "Put-call parity hold only for European options."

@Charly: thank you for your feedback. I agree about the ratio: I believe the ratio is not necessarily 0.75 of top 5%. I belive it is 0.XX (unknown ex ante) of top 5%.

@FoQ, Thanks for your support and feedback. I enjoyed the questions you posted to the forum, thanks for your engagement!

@NewTurtle. Wow, you have an awesome memory. Thank you so much for posting so much detail (I am definitely using this for future questions)

@Alex. Thank you for posting really helpful detail!
Re: Question on rating migration which I don’t remember as being part of the curriculum for level I
Technically, L1 introduced the migration matrix and L2 required a calculation, so "fair" would be a qualitative query for L1 and a calculation on L2.

Re: "I only wish there would be a way to align the level of difficult of the exam with the questions we candidate have access to."
Correct me if you disagree, but I don't feel like my challenge is difficulty per se, but rather (i) nature or type (e.g., wordy question) and (ii) topic/domain; my challenge is predicting what GARP will test on

Thanks, David

 

[troubleshooter]

@abl2117:
"- Question on rating migration which I don’t remember as being part of the curriculum for level I."

Actually rating migration is part of level I. It's in valutation and risk models...

 

[troubleshooter]

Level I contd..
----------------------------------------------------------------------------------------------------------
1. Cash Flow/Compounding question: You are investing for 30 years. You invest $10 today and 20$ 15 years from now. For the first 10 years the interest rate is x% with quarterly compounding and for the rest, its y% with semiannual compounding. At the end of investment, you have $100. y is given and you need to find x.

2. Hypothesis testing: What happens when you reduce the level of significance?
Combination of answers with increase or decrease in Type I and Type II errors.

3. LTCM - Which of the given answers is the issue problems with LTCM problem?
Answer was Liquidity due to unexpected market moves.

4. Issue with Metalgeselschaft - What is not the issue with MG? Quite straightforwad if you read the case. I think the choices where Fraud, Credit Risk and others that I am not remember but answer was quite straightforward.

5. CRO says there is the chance of making a billion or more of loss in a year is 99%. The next year 2 billion loss is realized and the bank goes bankrupt. Is it a risk management failure? Difficult question as the choices were hard to rule out.

6. Cheapest to deliver: Given a set of deliverable bonds for a futures contract. Upward sloping yield curve is given. Based on the coupon, you need to determine the cheapest to deliver. Explained well in Hull.

7. Perpetuity question - Given coupon rate paid semiannuly and given the value of the annuity. Need to calculate the yield to maturity.

8. 3 or 4 questions on optmal hedging ratio. Given Asset value to be hedged, contract size and price, standar deviation of the asset and the futures and correlation, calculate the number of contracts. This was overdone by GARP I thought but who's complaining?

9. Probability/Bond Price combination question: Given probability that a company's bond rating would migrate to 4 different rating classess next year along with the credit spreads for each of those ratings. Also given is the risk free rate. You need to find the price of the two year zero coupon bond issued by this company today with Face Value of 100. Find expect value of the spread add risk free rate. Use that to discount the Face Value for two years..

10. Basis Risk: If an asset is hedged by a forward contract on the same asset (with same volatility and correlation = 1), then would there be basis risk in it. I thought the answer was yes, as there was flexibility in the delivery location or something like that.

11. Commodity Derivative: If you using long on a commodity derivative based on the index of transportation, shipping and insurance cost as your hedge, what are you hedged against ? 1. Commodiy Price 2. Commodity Price and Shipping Cost 3. Shipping cost 4. Neither. I thought the answer was shipping cost only.

12. You want to hedge against volatile interest rates. What would you use:
A. Call Swaption (Payer Swaption)
B. Put Swption (Receiver Swaption)
C. Long Inter Rate Cap and Interst Rate Floor
D. Can't remember

Answer was A and C.

 

[cine]

@ New Turtle, that was awesome memory.
There is so much common between L1 and Full exam. I find that my swaption question has gone wrong but feel good as the rest seem correct

There was this question which really puts the motive of GARP straight. To test the conceptual understanding !

Q: There is a bond with coupon at 8% and yielding 6% a year. Time to maturity is 5 years. What happens to the price after one year?

While one could easily calculate using calci, one could easily make out that it's a premium bond and price should come down with time due to Pull-to-par effect.

Regarding "CRO says there is the chance of making a billion or less of loss in a year is 99%. The next year 2 billion loss is realized and the bank goes bankrupt. Is it a risk management failure? Difficult question as the choices were hard to rule out"

I think there were two close options. I went for "It shows risk management failure as they couldn't not predict loss of so high magnitude that was to happen only next year" or something close to this. The other one was sounding equally correct I guess....

 

[benoit16]

@benoit16, Thanks for your further input. About the transparency, I wholeheartedly agree, as a provider I totally support the value of transparency. But frankly, GARP will need to improve their own rigor before they can be transparent.
I must be dense, what happens on the 5th of January?
​

As far as I remember, the 5th is the result (happiness or sadness) day.

 

[shauncass]

Dear David,

A) One of my colleagues who has cleared the exam last year was telling me that there are weightages for each question as well..i.e. maybe an easy question could be worth only 0.5 marks and a tough question would be worth 2 marks...is this information correct?

As far I understood from the many readings online, every question from any section would have equal weightage...Pls clarify..

B) Also, as per your initial feedback from the participants, what is your rough estimate of the overall cutoff?

Regards and hope 44.14% clear the exam again

 

[Charly]

@ New Turtle, that was awesome memory.
There is so much common between L1 and Full exam. I find that my swaption question has gone wrong but feel good as the rest seem correct

There was this question which really puts the motive of GARP straight. To test the conceptual understanding !

Q: There is a bond with coupon at 8% and yielding 6% a year. Time to maturity is 5 years. What happens to the price after one year?

While one could easily calculate using calci, one could easily make out that it's a premium bond and price should come down with time due to Pull-to-par effect.

Regarding "CRO says there is the chance of making a billion or less of loss in a year is 99%. The next year 2 billion loss is realized and the bank goes bankrupt. Is it a risk management failure? Difficult question as the choices were hard to rule out"

I think there were two close options. I went for "It shows risk management failure as they couldn't not predict loss of so high magnitude that was to happen only next year" or something close to this. The other one was sounding equally correct I guess....​

Hello again!!!!

The CRO question was really crazy. Everything but one answer seemed correct. I finally chosed the one that said that with the information provided you could´t decide.....

According to the swaption question, I stated as the correct answer the Put Sawtion (as the one that has the right to pay fixed) following the Jorion.....Yesterdy night I read the forum and realised that newturtle and David discussed this issue reaching no conclusion. Let´s see what they do decide.....

 

[troubleshooter]

CRO says question is really not fair I thought. I also picked information provided was not enough. But it could go either way. VAR is not supposed to cover the lowest 1% and the loss of 2 b is not unusual. That way it's not risk management failure. Then we are both okay. But if it is assumed that he did everything, eg, stress testing, but based his conclusion upon VAR only (of course there was not enough information in the paper to rule each one out completely), then it is risk management failure as he failed to evaluate the 1% tail scenarios.

Regarding the swaption question, I think Call/Put is not that well defined for swaptions so GARP provided Payer or Receiver swaption in the parenthesis. I chose Payer swaption as it sounds like FIXED rate payer and the cap and floor. Who knows what's right and what's not ?

Looking back at the test, I don't think I would have done any better even if I put in 1000 hrs of extra studies. I did not think it was a fair test as it did not focus on many of the important areas, eg. Linda Allen, Michael Ong...

 

[David Harper CFA FRM]

@NewTurtle, thanks again for great detail! I had a few thoughts on a few of these recent questions, for what it's worth:

In regard to: “CRO says there is the chance of making a billion or less of loss in a year is 99%. The next year 2 billion loss is realized and the bank goes bankrupt. Is it a risk management failure? Difficult question as the choices were hard to rule out”

I am guessing this refers to Stulz' paper, Risk management failures: What are they and when do they happen? where his point is that a big loss is not prima facie evidence of risk management failure: "A large loss is not evidence of a risk management failure because a large loss can happen even if risk management is flawless ... the returns of LTCM do not tell us anything about whether its risk management failed"

In regard to: : There is a bond with coupon at 8% and yielding 6% a year. Time to maturity is 5 years. What happens to the price after one year?

IMO, that is a good question that can be answered quickly, just as Chinni shows: it is a premium-priced bond that will be pulled to par.

10. Basis Risk: If an asset is hedged by a forward contract on the same asset (with same volatility and correlation = 1), then would there be basis risk in it. I thought the answer was yes, as there was flexibility in the delivery location or something like that.

This question concerns me, and I am going to foward to GARP as potentially problematic. It appears to refer to Commodities and Commodity Derivatives, where, per the FORMULA for basis risk, basis risk =0: "This equation shows that basis risk is zero when variances between the Futures and spot prices are identical and the correlation coefficient between spot and Futures prices is equal to one."
i.e., the question appears to want a zero per the formula. But NewTurtle is absolutely correct: strictly speaking, we would almost never say there is no basis risk. The wording of the question requires great care, for a "novice answer" will be zero but a more "experienced answer" will give non-zero (e.g., historical correlation is time varying, delivery may not match as NewTurtle says)...I will definetely follow-up with GARP on this.

@shauncass:
A) no, GARP has never given various weights to questions. All question weigh equally.
B) I do not have enough information to estimate cutoff, apologies

Thanks,
David

 

[troubleshooter]

In regard to: : There is a bond with coupon at 8% and yielding 6% a year. Time to maturity is 5 years. What happens to the price after one year?

I also thought this was a great question. But to be perfectly honest, I have to admit "Pull to Par" concept never came to my mind while doing it under the pressure. Soon as I saw this question, I knew that I could do this in 30 sec. with my HP 12C.

 

[ankit015]

Quick question...multiple option can also be correct?....i dint know that?

 

[troubleshooter]

Not really. It is applicable when you have choices like.
A. i, ii
B. i, iii, iv
C. iii, iv
D. v only

You still have to select only one of A, B, C or D.

 

[[email protected]]

Hi David,

Thanks for the patience answering my queries during the preparation for the exam. These replies really help in getting a firm understanding of the subject.

The FRM questions are difficult. It tests the candidates time management, firm understanding and application of the subject, identify and avoid or diffuses the "tricks" embedded in the question.

It will save us a lot of time in preparation for the exam if the questions and replies in the forum could be filtered using the AIMS so that the candidates has an expectation of how the question will be asked on the topic thought the creativity of the examiner can not always be predicted. I think "historical simulation" (discussion on the past year question) will be very helpful here.

The existing practice questions and answer are helpful and therefore should be retained. But, more focus on the application aspect such as the candidate should expect " pull to maturity" will come into picture for a certain scenario. The impact of the parameters changes on the formula (such as BSM) and interaction between formula such as BSM and put call parity and the prices of the options.

Thanks
Learning

 

[gundu]

As I understand from all the above discussions & the FRM 2009 paper...GARP likes to test if a candidate can bring in logic into fuzzy situations...which is practically the RISK PROFESSION.

That is the reason of the so called famous "put call parity question with American options"...being tested...!

Many questions in the PM sessions had vague & close scenarios to choose the correct answer from...all this actually calls for the REAL APPLICATION of a candidated TRAINED in Risk Analytics & not any body who just crams up a few formulae 10-15 days before the exams.

Hence, I believe that more candidates with exposure in RISK PROFESSION is preferrred by GARP & so must be the case as JOB MARKET is not giving credit to a person with a FRM degree but no exposure in Risk.

So all said and done GARP is perfectly justified in creating the type of paper as seen by all of us on 21st November,2009.

I wish GARP should maintain its high level of difficulty that has been its tradition & not depart from it...for the rest us guys...we must just slog it out & work on our BASICS...thats the real MANTRA.....!!!

 

[David Harper CFA FRM]

@Learning,
Thanks for your engagement in the forum. I thought you asked some great questions (meaning, they helped me to learn, also).
In regard to "historical simulation," yes, I totally agree, I have started working to collect/tag helpful threads.
In regard to "more focus on the application aspect," and "the impact of parameters," I really do agree with that also ... I think we have to shift the emphasis more toward application. It is a *great* point!

@mita,
I agree with you, to a point. However, I disagree that an exam with a "a high level of difficulty" excuses fuzzy questions. Absolutely, the FRM should be difficult and reality-based, but this requires even more rigor than usual. If you look at past exam practice questions, there is occasionally too much imprecision and subjectivity required. (can we at least agree that three different approaches to surplus at risk--one in Jorion and two in the practice exams--is not good? or that we have two very different definitions of tracking error floating around?)

The put-call parity with American call option is, I suppose, debatable. I personally do not see how logic necessarily gets you to the final answer. I think the answer to this questions relies on whether you assume C=c or C >= c, and I think that decision is too expert for the exam (as I've suggested, C>=c seems quite reasonable unless you have time to conduct research). But, okay, I concede this is maybe a good example to support your point.

However, the question that looks to apply the basis risk formula and get zero basis risk (b/c cross volatility = 1.0 and correlation = 1.0), it seems to me, is a good example of a counter-example: here, you will get 0 by applying the formula (and that is the correct answer). However, from experience, we know basis risk is never zero...

...but i absolutely agree that the mantra is "slog it out and work on the basics." I think that is the only real solution!

David

 

[mdecav]

David,

Post-exam do you give/get direct feedback from GARP about the exams themselves?

 

[troubleshooter]

I don't know why but Box Spread really bothered. If they throw in stuff outside of AIMs to make the exam difficult, they should find other ways to be creative.

To be fair this question and the incomplete question on the probability tree should be thrown out of markings...

 

[amvantagev8]

Hi David,

I took the L1 FRM in Toronto and I found the exam quite difficult. The AM session went ok for me and I had 5 min to spare to check some of my answers. The PM session didn't go so well, because I got stuck on a couple of calculations and I had to guess quite a few answers, because I ran out of time.
I thought there were a lot of questions on GARCH and VAR on the exam and few on plain statistics. I guess GARP really wants you to apply Gujarati's theory to risk assignments.
In preparing for the exam, I did most of the practice questions on Bionic Turtle and the FRM practice exam. I still found the actual exam questions to be very different.
The "box spread" issue has already been raised in other posts. I don't understand why GARP put a question on box spreads in the exam, when they could have chosen any of the assigned option strategies.
I really enjoy the contents of Bionic Turtle and the webcasts, notes, spreadsheets and practice questions were a great help!
I'll keep using Bionic Turtle, because I suspect I have to redo L1 in May 2010. At least now I have a good impression of the way GARP tests your knowledge.

Thanks
Melle

 

[David Harper CFA FRM]

@mdecav:
I do give feedback; e.g., last year, I provided a detailed summary based on forum conversation, and I included a list of "issues" related to practice questions that we disputed (e.g., incorrect definitions of alpha or unassigned definitions of surplus at risk). And frankly, maybe because we let the practice questions "survive" the forum scrutiny (there really is no substitute for this kind of transparency: eventually, incorrect approaches are fixed), our track record on contested questions is nearly 100%. GARP basically agreed to my entire list last year...

And I am, again, currently collecting feedback for GARP, based on this year's forum threads and findings.
(Probably, I will just post a copy of the document here to the forum ...)

As far as GARP receiving feedback, I know they read the forum (and definitely this thread). Beyond that, I don't know their plan. Will GARP give feedback (e.g., like reporting back test results per provider), they did not indicate such a plan this year (2009 was unique in many respects).

@NewTurtle:
I suppose folks can debate this, but the box spread bothers me, too. I want to know the AIMs are a reliable "map" of the testable terrain; if they are not, what good are they and why am I going through them one at a time? I have told GARP this: I think the questions should all be vetted, as a final step, against the assignments. I guess in the case of the box spread, the answer can be: it's not on the AIM, but it's in the chapter...but, frankly, there is so much potential material like that. As a trainer, I want more help in where to focus and what to omit.

David