Level 2: Post what your remember here...

[David Harper CFA FRM]

Here's an attachment with the extract of the questions that were remembered by our posters here. Thanks to all for posting the questions. There was total of 76 out of 80 questions accounted for.

@troubleshooter, thank you for sharing the summary. The question of ethical post-exam sharing was earlier raised. I consider your particular post ethical and compliant with GARP's standard, as it does not replicate the entire Q&A in detail. We have received the official guidance. Please note that GARP is not opposed to exam postmortem discussion, as I was informed yesterday: "General recollections of exam questions is fine and I believe a valuable resource for candidates (and exam prep providers!)."

As a practical matter, a good way to help with compliance is to omit the four answers offered, this helps to avoid an exact "stem plus answer choices" replication.

GARP's official guidance:

Candidates can share general recollections of exam questions, which we believe to be a valuable resource for candidates and exam prep providers, but candidates should not share detailed replication of any exam question in any forum. At registration, candidates agree to abide by the Code of Conduct which states that they shall not engage in any conduct or commit any act that compromises the integrity or validity of the examinations. And on Exam day, all candidates sign a “Registrant Agreement” whereby it is written that all exam material is copyrighted and cannot be shared publically without GARP’s permission.

 

[David Harper CFA FRM]

David, There was only viable choices on this convertible arbitrage that asked when you get an increase in CF - 1. Coupon on the bond 2. Share price going down. So does you answer above mean it would be coupon?

Hi troubleshooter, It's not enough information for me: assuming a dynamic vanilla convertible, ANY of the sources of return could manifest as in increase in cash flow (Stowell's Figure 1 will demonstrate this, i would think). Wording matters (if the convertible bond position doesn't change, the coupon cash inflows would not increase any more than fixed coupon cash flows don't change, so I'm not understanding the exact meaning of "increase"). If the share price goes down, the dynamic arb trade is buy shares (reduce the short share position) which is an outflow. So, neglecting dividends, I think the basic dynamic of "share pricing going down" is net M2M gain but cash outflow = zero inflow (due to unrealized capital gain on short) - outflow to purchase shares. "An increase in cash flow" is insufficient for me .... my initial reaction is neither creates an increase in cash flow (why would the coupon increase? share price down implies net cash outflow but capital gain) but it requires assumptions ... sorry, thanks,

 

[troubleshooter]

I agree. I guess it's one of those badly worded questions where it can interpreted in many different ways. Your luck rests on your interpretation. I will not decisively blame GARP for this as I did for the DV01 gap question... Anyway, thanks for taking your time... And thanks for getting GARP's guidance on post exam discussions. Seriously, I did not want to get into trouble for discussing the exam in a public forum...

 

[David Harper CFA FRM]

Sure thing - I forwarded feedback on the "DV01 gap" question to GARP directly, on its own, to three of my contacts at GARP. I wrote them "I would find it hard to believe it possible this error could get to an actual exam." Yet it appears possible. (officially, they won't confirm as that would open a can of worms.). As much as i would love to see the actual exam, i have come to understand why they cannot publish it, and I am almost relieved they won't.

On this commitment to quality, all i can say is here is how i feel about it:
although the order might be: (I was mad and confused for a while [how is this possible?] ... now i'm just sad and concerned)

 

[jdg123]

On the Liquidity Duration Question. Given a position in one security with an average daily volume as well as a position, it was easy to solve for the liquidity duration of each. However, in the answers, I might've chosen the maximum of the two. For instance, if one position takes 4 days to liquidate and the other 8, I would think it would take 8 days to liquidate the portfolio. However, the only answers feasible were 6 or 12. Most people on this forum chose 6, as the average of the two. However, I chose the sum of them, as the average really doesn't make much sense. If one has 200 securities and 199 of them could be liquidated in a day and the remaining one takes a year, I would not think the portfolio's liquidity duration is the average of the 200. However, I could not find the definition of the liquidity duration of a portfolio. As my logical (max of an individual position) was not a choice, I thought sum made more sense than the average. Any opinions?

 

[troubleshooter]

On the Liquidity Duration Question. Given a position in one security with an average daily volume as well as a position, it was easy to solve for the liquidity duration of each. However, in the answers, I might've chosen the maximum of the two. For instance, if one position takes 4 days to liquidate and the other 8, I would think it would take 8 days to liquidate the portfolio. However, the only answers feasible were 6 or 12. Most people on this forum chose 6, as the average of the two. However, I chose the sum of them, as the average really doesn't make much sense. If one has 200 securities and 199 of them could be liquidated in a day and the remaining one takes a year, I would not think the portfolio's liquidity duration is the average of the 200. However, I could not find the definition of the liquidity duration of a portfolio. As my logical (max of an individual position) was not a choice, I thought sum made more sense than the average. Any opinions?

Max definitely makes sense. Sum does not make sense as you do not have to wait for one of them to be fully liquidated before you begin with the other one. I think LD came out as 4.xx and 5.xx. The choices were at 5, 10, and higher than 10 values. I would think 5 was the best answer.

 

[David Harper CFA FRM]

@jdg123 - I am aware of no definition for "liquidity duration" in any of the assignments. Admittedly, I've only read them a few dozen times, so maybe i missed it but even Google doesn't give me solid references. I'd appreciate any legitimate reference ... I can see that it sounds like an average days to liquidate or something ... (it sounds to me like an abuse of "duration" frankly)

 

[jdg123]

The question did not say average, otherwise it would have been very easy. The reason I did not take the average is because it does not make sense to me. Now, the sum does not either. However, if one were only liquidating one at a time, the answer could be the sum. I think the numbers were as troubleshooter said, however, I think the 5.xx was closer to 6 than 5, so I just couldn't think of any way to get 5, other than to take the average, which just didn't make sense. So, at least the sum came to approximately 10, so I figured it was one of the choices. Anyway, I guess we'll never know. I did not look back at the original readings. Perhaps if I read it two dozen times, I will be able to find the answer. I'm not sure how the question writers can write such questions where there is no real way to find the answer. In this case, we all know how to calculate the liquidity duration of an individual position. For a portfolio, I don't know, to me, if you could (and I don't see why you couldn't) liquidate both on the same days, the answer, logically, would be the max of the two (which was not one of the choices). Frustrating.

 

[jdg123]

I still have not found a convincing answer on the Q-Q plot of a normal vs. an empirical distribution, or something like that. For some reason, I was thinking of empirical as being a lognormal distribution, but thinking about it now, couldn't it be any distribution. I couldn't logically figure out how points could all be above or below. That's the only reason why I chose one that started out below and ended up above, though I don't think that is how a lognormal distribution would look.

 

[jdg123]

On the Madoff, I thought it was obvious that the consistency of returns was a red flag. However, others have written other things here.

 

[jdg123]

Also, when calculating VaR for options, do we treat deep in the money calls the same as the actual stock but exclude out of the money calls completely?

 

[jdg123]

I did not think data problems led to different VaR measures. As historical just looks at history and the others have built in assumptions, I thought it was obvious that it was differences in assumptions that led to different VaRs. I'm not sure why others thought it was data problems.

 

[troubleshooter]

@jdg123:
1) Liquidity duration: I guess you never know about this one... Taking average makes sense if we are asked average liquidity duration but I doubt if "average" was used in the question, otherwise it will be quite straight forward and we would not be talking about it now. Sum may also makes sense if the question writer has assumed that you could not liquidate the 2 positions at the same time but ignored to disclose that assumption to the test takers...
2) Q-Q Plot I thought was fairly strait forward. I chose the third picture where the empirical was straight to left of the y axis and then over the normal line to the right showing extreme values at the upside only.
3) Madoff was very straight forward. Except for consistency in returns, nothing else made much sense.
4) Calculating VARs for options, the delta for deep out of the money call = 0 and delta for deep in the money call = 1 and delta for futures = 1, so you don't include deep out the money call. This was in one of the GARP practice exam too...
5) Different VAR measures was quite badly written question too... The difference was not that significant and I would expect the difference as valid random outcomes. I think it was something like 245,000, 250,000 and 275,000 But we had to pick one of the choices so I chose different assumption, while Data error can always produce such errors. So this one is up in the air too.

 

[shanlane]

David,
There was only viable choices on this convertible arbitrage that asked when you get an increase in CF - 1. Coupon on the bond 2. Share price going down. So does you answer above mean it would be coupon?

Anyone who thinks this question was had NO good answers, please email GARP, as I did.

I happened to put stock price going down as the correct answer because I didnt think that ANY of the answers were correct, because NONE of them led to an INCREASE in cash flows. The bond coupon was a POSITIVE cash flow but not an INCREASE in cash flows. The stock price going down leads to an increase in PROFIT.

NEITHER CHOICE CORRECTLY ANSWERS THIS QUESTION!!!

Instead of reading trying to read GARPs mind, let them know that they asked a bad question and that they either need to not count this question or count both answers as correct.

[email protected]

Shannon

 

[troubleshooter]

Anyone who thinks this question was had NO good answers, please email GARP, as I did.

I happened to put stock price going down as the correct answer because I didnt think that ANY of the answers were correct, because NONE of them led to an INCREASE in cash flows. The bond coupon was a POSITIVE cash flow but not an INCREASE in cash flows. The stock price going down leads to an increase in PROFIT.

NEITHER CHOICE CORRECTLY ANSWERS THIS QUESTION!!!

Instead of reading trying to read GARPs mind, let them know that they asked a bad question and that they either need to not count this question or count both answers as correct.

[email protected]

Shannon

Hi Shannon:
Done that already. Back in 2009 when I did Lvl 1, I did the same and I got a response saying if they agree that the question is flawed, they will be excluded... Here's copy of my communication...


First Email
________________________________________
Sent: Monday, November 23, 2009 8:58 AM
To: FRM
Subject: Level I Exam Question Issues
I am a level I candidate and took the exam in Toronto, Canada on Saturday. I would like to report a couple of issues I noticed with the exam questions.

1. There was a question on box spread. Box spread, though in John Hull textbook, is clearly not in curriculum as per FRM AIMs.
2. There was a question on probability tree that was incomplete. At the last part of the question, it did not specify which way the tree went.

I am sure GARP has processes in place to avoid such operational risk exposures. I would like to know how GARP will go about marking these questions - whether they will be excluded from the markings or they would be treated as any other questions. Also, can we expect questions not in AIMs to appear in Level II in May?

Thanks,
Garp Response 1:
________________________________________
From: Chris Donohue [mailto:[email protected]]
Sent: Friday, December 04, 2009 9:47 AM
Subject: FW: Level I Exam Question Issues
Thank you for your message and your interest in the FRM program. Here are my responses to your questions:

1. The AIMS do have: "Describe and explain the use and payoff functions of spread strategies" and then gives some examples. Because box spreads were not listed in the examples does not imply that they are not part of the curriculum. They are covered in the reading.
2. After the exam, we always get feedback from candidates about potentially problematic questions. We check each of these questions closely and if we agree that there is a problem with the questions, we drop it from the exam scoring. I will make sure we check the question you referenced. Thanks for pointing this out.
As we write the exam, we reference each question to an AIMS statement and will do so for the 2010 exams as well.
Thanks again for your interest. If you have any additional questions, please don't hesitate to contact me directly.
Best regards,
Chris

Christopher J. Donohue, Ph.D.
Managing Director, GARP Research Center
Global Association of Risk Professionals (GARP)
111 Town Square Place, Suite 1215
Jersey City, NJ 07310 USA
t: +1-201-719-7261
m: +1-908-656-0961
e: [email protected]
www.garp.org
Second Email:
___________________________________________________
Sent: Friday, December 04, 2009 9:52 AM
To: Chris Donohue
Subject: RE: Level I Exam Question Issues
Chris:
Thanks for the response. For my Level II preparations, I guess this would mean that I should take the assigned readings in whole as the testable material and regard AIMs as only a guidance and not as an exhaustive and explicit listing of testable material. Would this be correct?
Thanks,

Garp Response 2:
_____________________________________________________
From: Chris Donohue [mailto:[email protected]]
Sent: Friday, December 04, 2009 10:29 AM
Subject: RE: Level I Exam Question Issues
Yes, I would view all the reading material as testable unless we have explicitly said that a section will not be tested, and use the AIMS to guide your studying. I can say confidently that if you are comfortable with the AIMS, you will not have trouble passing this exam even if there are a few questions that are loosely connected to the AIMS. We have pushed the exam development in this direction over the last few years and will continue to do so.

 

[shanlane]

Hi Shannon:
Done that already. Back in 2009 when I did Lvl 1, I did the same and I got a response saying if they agree that the question is flawed, they will be excluded... Here's copy of my communication...


First Email
________________________________________
Sent: Monday, November 23, 2009 8:58 AM
To: FRM
Subject: Level I Exam Question Issues
I am a level I candidate and took the exam in Toronto, Canada on Saturday. I would like to report a couple of issues I noticed with the exam questions.

1. There was a question on box spread. Box spread, though in John Hull textbook, is clearly not in curriculum as per FRM AIMs.
2. There was a question on probability tree that was incomplete. At the last part of the question, it did not specify which way the tree went.

I am sure GARP has processes in place to avoid such operational risk exposures. I would like to know how GARP will go about marking these questions - whether they will be excluded from the markings or they would be treated as any other questions. Also, can we expect questions not in AIMs to appear in Level II in May?

Thanks,
Garp Response 1:
________________________________________
From: Chris Donohue [mailto:[email protected]]
Sent: Friday, December 04, 2009 9:47 AM
Subject: FW: Level I Exam Question Issues
Thank you for your message and your interest in the FRM program. Here are my responses to your questions:

1. The AIMS do have: "Describe and explain the use and payoff functions of spread strategies" and then gives some examples. Because box spreads were not listed in the examples does not imply that they are not part of the curriculum. They are covered in the reading.
2. After the exam, we always get feedback from candidates about potentially problematic questions. We check each of these questions closely and if we agree that there is a problem with the questions, we drop it from the exam scoring. I will make sure we check the question you referenced. Thanks for pointing this out.
As we write the exam, we reference each question to an AIMS statement and will do so for the 2010 exams as well.
Thanks again for your interest. If you have any additional questions, please don't hesitate to contact me directly.
Best regards,
Chris

Christopher J. Donohue, Ph.D.
Managing Director, GARP Research Center
Global Association of Risk Professionals (GARP)
111 Town Square Place, Suite 1215
Jersey City, NJ 07310 USA
t: +1-201-719-7261
m: +1-908-656-0961
e: [email protected]
www.garp.org
Second Email:
___________________________________________________
Sent: Friday, December 04, 2009 9:52 AM
To: Chris Donohue
Subject: RE: Level I Exam Question Issues
Chris:
Thanks for the response. For my Level II preparations, I guess this would mean that I should take the assigned readings in whole as the testable material and regard AIMs as only a guidance and not as an exhaustive and explicit listing of testable material. Would this be correct?
Thanks,

Garp Response 2:
_____________________________________________________
From: Chris Donohue [mailto:[email protected]]
Sent: Friday, December 04, 2009 10:29 AM
Subject: RE: Level I Exam Question Issues
Yes, I would view all the reading material as testable unless we have explicitly said that a section will not be tested, and use the AIMS to guide your studying. I can say confidently that if you are comfortable with the AIMS, you will not have trouble passing this exam even if there are a few questions that are loosely connected to the AIMS. We have pushed the exam development in this direction over the last few years and will continue to do so.

Interesting. I will make sure I email them about other problematic questions as well.

Thank you!

Shannon

 

[FRM_Exam]

So I am getting an exact 60 out of 80 (after considering that I have got wrong - (1) anything which I am unsure about I mark wrong; (2) the answers I know I have surely got incorrect now after reading the reasonings over here)

I hope it is enough for me to pass the exam - I just can't think of re-taking the exam.

Does anyone know what might be the average score with this kind of an exam? I understand it depends on how well the top 5% ppl do and then a specific percentage of that top 5% score - that specific percentage is the most crucial

 

[majesta]

I tried to register for november 2012 Part II exam, I am really shocked because it gives me access to register. What does this mean? maybe I failed but I did good

 

[FRM_Exam]

Hi Majesta, Why are you trying to register so early - ofcourse it will allow you to register - I think the trick only works 1 day before or maybe the result day - they wouldn't have updated the systems yet. Or maybe they will have even fixed the bug this year - so no guarantee it will work

 

[shanlane]

Quick update. I bet they say this every year, but I contacted GARP and got a few responses about the quality and consistency (or lack thereof) of the level 2 exam and they have been pretty good about getting back to me.

I just wanted my voice heard and to let them know that some of the questions had serious flaws. They said that they are aware of it and will try to make it so that nobody is adversely affected by bad questions.

I guess what I am trying to say is that with only a week or so before the results are released, if you have any comments or remember any bad questions it might be worth shooting an email to the powers that be over at GARP.

I wish everyone the best of luck!

Shannon