Exam Feedback November 2015 Part 1 FRM Exam Feedback

[Moey]

Thanks all [re sharpe clarification] - another incorrect easy one. :/

How about the futures open interest question? Think the question was 100 open long, 50 open short, and 50 close sell - what's the volume [change] (100 or 0) and open interest (either 0, 50, 100)? I answered 100/50.

 

[mishrasa]

Thanks mate! Made my day.

All the best you too!
Cheers!

@mishrasa

I think that you do have a chance of passing with 50 - 60%!
There are a couple of people who only studied one week for the Part I Exam and then they passed, even with a quite good quartile ranking.
In any case, I wish you the best of luck and will keep my fingers crossed for all of us for January 5th!

 

[vistag]

For anyone interested, these are the questions I can remember:

Is it not a violation of GARP code doing this? Surprised this has come from S666 !!! Discussing paper difficulty is different and posting exam question information is different.

 

[S666]

Hey @vistag ...you're probably right actually - I think my post verged on giving away too much detailed info and hence I have deleted it. I had seen many posts on here discussing specific questions from previous exams and therefore was under the impression that it was fine to discuss questions in detail - I had read somewhere before that this was actually one of the main differences between the CFAI code of ethics and the GARP code.

Searching around for more info, apparently you can discuss particular questions in general but replicating question/answer stems is close to crossing a line. I'll err on the side of caution and remove my post.

However if you search the exam feedback threads from past exam dates, you will see that people discussing questions in quite some detail is nothing new.

 

[MRWUSH]

I participated at the Part 1 exam.

 

[heretolearn]

I found the exam very long - not too hard quantitatively, but long. I had to guess quite a few questions that required calculations in the end even though I think I could have solved them with more time. I also remember a lot of CAPM, VaR, binominals, swaps, FX, and a lot of qualitative questions. I am glad it wasn't all math, it would have made it even harder for me to finish on time.

Dose anyone know when we can expect results?

 

[Numerical Wizard]

@heretolearn
You can expect the official results on January 5th 2016!
I have received an official email from GARP in which they have said official results will be out on that day.
Good luck to you!

I found the exam very long - not too hard quantitatively, but long. I had to guess quite a few questions that required calculations in the end even though I think I could have solved them with more time. I also remember a lot of CAPM, VaR, binominals, swaps, FX, and a lot of qualitative questions. I am glad it wasn't all math, it would have made it even harder for me to finish on time.

Dose anyone know when we can expect results?

 

[heretolearn]

Thank you @Numerical Wizard! Same to you

 

[bplloyd919]

Felt somewhat easy although that makes me nervous...

One tricky question about calculating VaR on a delta neutral portfolio using the delta normal method which was the only question I had to change my answer to, as it took a double take to realize that this meant that VaR is zero! Can anyone confirm that this is correct?

I know the question you are referring to, however I disagree that the answer was zero. Zero was the "trap" answer I believe: the delta-gamma method is an approximation. It is the same as a second-order Taylor approximation. Therefore, being delta-gamma neutral only means you are approximately neutral, thus there is no way VaR can be zero. The correct answer I believe was that VaR is greater than zero but less than x (whatever x was, I can't remember).

 

[bplloyd919]

I thought the exam was tough, and I regretted not studying more, even though I was not unprepared either. The best way to explain how I feel about the exam is that I took for granted some things I thought I already knew pretty well and told myself I will be able to figure it out during the exam, instead of drilling it into my brain like second nature. Indeed, I could have figured those things out if I had 10 minutes per question. But under stress and rushed, it just was not happening.

I don't feel good about it. I won't be surprised if I did not pass. I believe I will slip by with the lowest passing grade possible.

 

[Moey]

Zero was the "trap" answer I believe: the delta-gamma method is an approximation.

You're right - I thought this too but didn't have time to overthink the "trap" answer of zero.

"See Market Risk Analysis, Value at Risk Models" by "Carol Alexander", page 273: https://books.google.com/books?id=e...TAA#v=onepage&q="the var is not zero"&f=false

 

[The Great Khan]

I know the question you are referring to, however I disagree that the answer was zero. Zero was the "trap" answer I believe: the delta-gamma method is an approximation. It is the same as a second-order Taylor approximation. Therefore, being delta-gamma neutral only means you are approximately neutral, thus there is no way VaR can be zero. The correct answer I believe was that VaR is greater than zero but less than x (whatever x was, I can't remember).

The question specified - 'using the delta normal method, what is the var', so they are looking for the approximated value which is zero.

It wouldn't be very valid for the answer to be anything else, as depending on the distribution any of the answers could be the true VaR. There are many ways that it could be >15k var, if for example it was a portfolio with a really long time to expiration and the first derivative of gamma is large.

Or it could trivially be a true var of zero if the delta neutral portfolio was cash

 

[The Great Khan]

What is the solution to the key rates question? I couldn't figure out how to solve that one.

 

[bplloyd919]

The question specified - 'using the delta normal method, what is the var', so they are looking for the approximated value which is zero.

It wouldn't be very valid for the answer to be anything else, as depending on the distribution any of the answers could be the true VaR. There are many ways that it could be >15k var, if for example it was a portfolio with a really long time to expiration and the first derivative of gamma is large.

Or it could trivially be a true var of zero if the delta neutral portfolio was cash

Hmm, you may be correct but I remember the question being worded something like "the portfolio was constructed so that it is delta gamma neutral.. what is the actual VaR?" Of course, if the portfolio is somehow quadratic (or linear) then YES - the VaR is actually zero. Else it is most certainly not.

I'm not here claiming my answer was correct as GARP chooses to score it. But I feel my answer was the correct choice in the real world, and that makes me feel good about that question given the practical slant that exists in the FRM program.

 

[bplloyd919]

What is the solution to the key rates question? I couldn't figure out how to solve that one.

Are you referring to the question that asked for the percentages in the replicating portfolio based on two bonds, given their respective Key rate exposures to the 2 yr & 5 yr? I can't remember my exact answer, but it was one of the options that involved a very small percentage to one, and very large to the other (the 4 choices were really two choices, with L/S flip flopped for the hedging positions). I'm not even certain I was correct, due to the nuances of the question, but I was able to match one of the answers by solving a linear system in 2 unknowns.

I think there were other Key Rate questions that I skipped and didn't have time to revisit (beyond filling in a random bubble on the answer sheet).

 

[S666]

The question specified - 'using the delta normal method, what is the var', so they are looking for the approximated value which is zero.

It wouldn't be very valid for the answer to be anything else, as depending on the distribution any of the answers could be the true VaR. There are many ways that it could be >15k var, if for example it was a portfolio with a really long time to expiration and the first derivative of gamma is large.

Or it could trivially be a true var of zero if the delta neutral portfolio was cash

I agree with this....the questions said "using the delta normal method, what is the VaR of a portfolio which has been constructed to be both delta and gamma neutral".

The delta normal method is an approximation in itself , calculated using either duration and convexity (for bonds) or delta and gamma (for options). So using the approximation to calculate VaR it is zero.

 

[S666]

Are you referring to the question that asked for the percentages in the replicating portfolio based on two bonds, given their respective Key rate exposures to the 2 yr & 5 yr? I can't remember my exact answer, but it was one of the options that involved a very small percentage to one, and very large to the other (the 4 choices were really two choices, with L/S flip flopped for the hedging positions). I'm not even certain I was correct, due to the nuances of the question, but I was able to match one of the answers by solving a linear system in 2 unknowns.

I think there were other Key Rate questions that I skipped and didn't have time to revisit (beyond filling in a random bubble on the answer sheet).

I may be wrong but I believe the question you are taking about was the replication of the 3% 1 year treasury bond using positions in a 5% 6 month maturity bond and a 4% 1 year maturity bond (all semi annual coupons).

The answer to that was short 0.005% of the 6 month 5% coupon and long 99.5% of the 1 year 4% coupon bond.

I believe the question the Great Khan is talking about involved key rate buckets of 0-2 and 2-5 years. I think there were actually 2 key rate questions, and I can't really remember what either of them asked exactly but I remember having a little play around and finding an answer that matched one of the available choices for both questions. That of course doesn't in any way mean I got them right as they can have legitimate looking wrong answer options, but I felt pretty ok about them at the time I remember.

 

[bil8]

Guys see after discussing , everyone has different opinion about few questions and everyone is justifying through their own logic based on study material.....This is what these professional exams like cfa and Garp are famous for ....They construct questions in such a way that two of the choices can be right at the same time However as far as exam was concerned or whether it was easy or difficult ...We should not judge it on the basis of people present here on this forum....Most of the people over here were well prepared and if we decide to calculate MPS on ths basis of people present here, it will be biased towards on the higher side....i still believe MPS will be around 60 considering the impact of overall population.............................What your thoughts about this

 

[mfonteneau]

You're right - I thought this too but didn't have time to overthink the "trap" answer of zero.

"See Market Risk Analysis, Value at Risk Models" by "Carol Alexander", page 273: https://books.google.com/books?id=e03cHzItv0AC&pg=PA273&dq="the+var+is+not+zero"&hl=en&sa=X&ved=0ahUKEwjBxqS-86fJAhVFChoKHQWnDdEQ6AEIHTAA#v=onepage&q="the var is not zero"&f=false

I'll chime in with this. I thought this too - however, the example in your link is for a 10-day VaR, which would require constant rebalancing over a 10-day period to remain delta-gamma neutral, and as such, the VaR value would be different from zero. I'm not so sure the exam referenced a 10-day VaR.

If you read further in your example, daily rebalancing would produce a VaR of zero.

Then again, the example also includes being vega neutral. Which wasn't touched upon in the curriculum (I don't think at least), so long story short, I have no friggin' clue.

 

[vistag]

Guys see after discussing , everyone has different opinion about few questions and everyone is justifying through their own logic based on study material.....This is what these professional exams like cfa and Garp are famous for ....They construct questions in such a way that two of the choices can be right at the same time However as far as exam was concerned or whether it was easy or difficult ...We should not judge it on the basis of people present here on this forum....Most of the people over here were well prepared and if we decide to calculate MPS on ths basis of people present here, it will be biased towards on the higher side....i still believe MPS will be around 60 considering the impact of out verall population.............................What your thoughts about this

After completing the exam even i was overjoyed. Primarily stems for the fact i was aiming to get around 50 correct , but managed to attempt 75 within the time (I predict my final score wd be around 55-60). Infact i spent lots of time double checking reading the question to make sure i was not being tricked withe easy looking question and ambiguous answer which wasted majority of my time. (Although i admit this approach helped in solving couple of ES related problems accurately)

To make things worse , my calci prcision has been set to 4 digits , as i didnt expect the need to go beyond tht (well served in mock exams), but there are atleast 2 questions where the precision went till 6 digits.(and oops forgot the precision set up input- cost me 1question for sure)

I agree the people in the fourm here are a biased sample. But most people in my Examintion center whom i probed were extremely confident immediately after writing the exam.

For exampe , i asked 2 guys right after i came out of examination room. One person said he attempted 95 and he is expecting 80 correct and was happy , the other one said the exam was difficult as he attempted only 90 questions and he expect 75 to be correct .

Thats when the lightening stuck my when i realised i completed only 75 questions which would translate to 55 total questions which can be correct and thats how the joy has fallen flat.

Then i see S666 and the legion of people ripping the forums apart which confirmed my fears

The bottom line is MPS might be higher , so am already out of the race