What's new

FRM Part 1 : Exam Done in Dubai 2011

LankyLint

Member
Haha yea. I suppose it is not. Tell me if you remember. It could be that you misread correlation to covariance or something. I don't think any question was wrong.

Can't wait for the results! I see that most people are saying the paper was easy but they either couldn't finish the paper or just didn't focus enough.
 

Meredius

New Member
As it's almost 2pm EST GTM -5 I think no on would read this before sitting for exam 1.

Also this is my first post. I was a bit short on time these past months but the videos and PRACTICE QUESTIONS /ANSWERS were EXTREMELY beneficial.

I did my exam this morning in Montreal. I don't feel "too" bad about it but I know that I would've liked (should've!) to spend another 20-30 hours doing some exercises.... it's hard to know WHAT they'll test though but it definitely was quantitative all the way. Still it naggles me that I knew where to get the info or where the missing formula is color coded in my head but doesn't want to materialize... though what does sadden me is I have a VERY good memory for qualitative information. Much moreso than formula's. So i'm thinking Level 2 should be more up my alley whence I cross that bridge....

There were 2 ethics questions as far as I remember...
-one which asked when do we become binded to the rules of Conduct? (when we sign up as member, register or whatnot, when we paass an exam, after 2 years of working... )
-also a trick one I think which had the 2 statements of how to conduct yourself.
I as a risk manager you should hold your employees and underlings to the ethics and professionnalism of the designation
II As a senior risk manager, you should delegate the responsibility of upholding ethics to your employees (hah delegate!)

There were 2-3 questions about correlation / covariance.
There were 2 one-step binomial delta-neutral pricing questions (one steppers!) with dividends
There were 2 put-call parity questions - again with dividends
There were 3 or 4 about neutral delta-hedging with option strategies (found this overkill a little)

And , yes David (couldn't help myself! but since I listened to you talk for 100 hours already, I feel like I know you somehow lol) there was one question about the rating agencies matrix... asked which "investment" grade bond has the least chance of downgrading over the year.

One odd one was comparing VaR and ES where it said VaR couldnt' drift with time because it was "spectral" (?)...

There were 3 questions regarding to graphs! 2 of them had 4 choices where they'd expose a type of investment correlated with another and you had to choose which curve of linear relationship best described it.

And no Black-Sholes....? Aside from one related to the greeks and their impact on option sensitivities.

Also on trick one which was pushing for sub-additivity on the VaR and another measure of fit (which is a no-no!)

I wish everyone the best. I'll also try to participate more in the forums as all the material available for 250$ is a steal and worth many times over it's price. I like how the pratice questions drill down more into the guts of a concept.

Kind regards,
Sacha

P.S. I discovered your RSS feed too late. I kept seeing daily questions in the emails and wondered where they were... I only found it 3 weeks ago. :/
Display it PROULDY in a link or banner!!
 

Hend Abuenein

Active Member
Wow, great memory Sacha, you managed to remember all this,

Maybe that's because you wrote this just after you left the exam?
I dropped to sleep the minute I reached home, since I didn't sleep for a minute the night before, and it seems I lost memory of all question specifics in that sleep.

I remember now the questions you mentioned.

There was one particular question I stopped at:
Bankers Trust was sued by Proctor & Gamble and -----Greetings, among the allegations against it:
I- The payoffs of the options it contracted for its clients were not clear to them
II- BT employees bragged about fooling clients.
Correct me if I'm wrong, but II was how I was proved true in court, not an allegation made against BT. So only I is correct. Right?

When I did practice questions on ethics, I always got it wrong, but the ones in the exam were pretty easy, I think.
You don't delegate your ethical responsibilities, and you become bound by code of conduct upon registration for FRM exam or GARP membership whichever comes first. (I'll always have doubts)

Good luck
 

Suzanne Evans

Well-Known Member
So where's David? packing for thanksgiving trip to family I suppose? Thought he'd be eager to get our feedback

Hi Hend,
Yes, David is probably preparing for the Thanksgiving Holiday. We are very eager for feedback. There is no doubt that David will login later to check for feedback. He is probably giving everyone a chance to take the exam and will catch up later this evening. For what it's worth, I'm following the forum and appreciate all of the feedback (as I know David does too!).
Thanks,
Suzanne Evans
 

Dmitrij

Member
So, I took part I and part II today. Part I went fairly well, although I had a few issues with it: I barely managed to get everything done in time, despite being fairly well prepared. Just simply reading/writing down everything takes a lot of time. In my opinion GARP should've either given 15 minutes extra or reduce the amount of questions by five. Then I was a little bit disappointed by the fact that there was not a single question asking us to compute the Black-Scholes model. Some questions were ambiguous, such as the one regarding delegation of responsibility. (I did actually recognise the 'delegate' part, and started thinking about the semantics of the word. I accepted it as true. That shouldn't be the purpose of the exam. by the way, not sure if this was part I or part II)

Some questions were notoriously tricky while some were so ridiculously easy that I had to look over it like five times to make sure I didn't miss anything. Plugging in variables, really? GARP should either group questions by type or by difficulty - it is very confusing to have to do a very difficult mathematical calculation and then to have something extremely simple after it. Also I expected more risk-neutral pricing and binomial trees and Greeks. Really sad that they weren't tested more. However, I was extremely happy that they didn't ask bond date conversions. Was also disappointed by the fact that no option strategies were asked.

Regarding part II: part II was pretty focused on VAR estimation and there were many qualitative questions. Also, nothing was every asked about knock-in, knock-out, Bermudan, compound etc etc options. It seems that this exam was light on options and more heavy on tail risk. I'm a bit disappointed, but maybe that's more because I'm a derivatives rather than a VaR type of person. Also expected a bit more on the structured finance part, such as structure of a CDO/CMO, difference between types of mortgages etc.

What I basically noticed is that there was very much Jorion and Allen, while the focus on John Hull was fairly low.

This is just a rough impression. I woke up at 6 AM this morning and spent the entire day doing exams wil 6.15 PM. Excuse me if this message is not very coherent, but it's the best I can do right now.

David, I will of course write a very good review, but it's conditional on me passing part I and II ;) Also of course I will offer feedback for BT etc etc. I have some more ideas that I will post later on, once my brain unmushes.

Regards,

Dmitrij
 

David Harper CFA FRM

David Harper CFA FRM
Staff member
Subscriber
Hi Hend: I took the morning off (to exercise and shop a bit; while you suffered, I was having fun :p), but I thank you for the update ...
Re: first page of information ("all rates are annually discounted unless question says otherwise"). Good to know! I will take credit for that, if you don't mind, I have bugged GARP repeatedly with links to confusions around compound frequency, so I am thrilled they added this standardizing point!!

@Sachal: Awesome detailed feedback, thank you so much for taking the time to share.
Re: "PRACTICE QUESTIONS /ANSWERS were EXTREMELY beneficial." Thanks, yes, going to emphasize those EVEN more (e.g., mock exams)
Re: rating agencies matrix. Ha, good to know!
Re: And no Black-Sholes. What??!! Hard to believe ....
Re: "I discovered your RSS feed too late. I kept seeing daily questions in the emails and wondered where they were " Okay, THANK YOU, yes I am linking to the feedback for our (slight) re-design. I totally agree, that's unforgiveable that it's not obvious to you

@Dmitrij:
Thank you for taking the time to write so much after your long day.
Re: "I was a little bit disappointed by the fact that there was not a single question asking us to compute the Black-Scholes model" Wow, I am too, considering that Merton (via Stulz) is such a strong theme, I am disappointed, I don't think they've ever omitted BSM
Re: "Also of course I will offer feedback for BT etc etc. I have some more ideas that I will post later on, once my brain unmushes."
Please do so, as you corrected several of my PQ mistakes, I am keenly interested in your feedback, Dmitrij!

Thanks, David
 

ebb

New Member
I am angered by the omission of Black Scholes. I went out of my way to memorize every single formula and concept in Black Scholes and that took the majority of my rote memory that I could have saved for other minute details.

There was one formula that bothered me in the exam majorly.

There are two assets, A and B. Variance of asset A denoted by VAR(A) = 0.25, Variance of asset B denoted by VAR(B) = 0.04. They are in equal weights in a portfolio. VAR(A+B) = 0.30. Calculate the correlation.

I thought I am indifferent to using w(A)^2*VAR(A) + w(B)^2*VAR(B) + 2 * (WA) * (WB) * Covariance(A,B) and isolate correlation accordingly. When I use this method, i get 4.55 which was not one of the answers.

While using VAR(A+B) = VAR(A) + VAR(B) + 2* Covariance(A,B) leads to an answer of 0.05 which is one of the answers. Can you shed some light here?

There was another question about using COV(AX+BX, BY+CY) like in the practice exam but I felt they lacked some inputs to properly solve it. At least I didn't bother spending too much time with it.
 

furqan

New Member
I felt it was tricky but not impossible my problem was time I think I was just too slow. I think question 22 was the one without weights and gave you covariance. I simply used the variance rule without the weights and got an answer that was listed but there was another in which weights were given.
 

LankyLint

Member
Bunnyblaster,

Var(A+B) = Var(A) + Var(B) + 2*Cov(A,B)

0.3 = 0.25 + 0.04 + 2 * Cov(A,B)

Cov = 0.01/2 = 0.005

Correlation = 0.005/(rootOf(0.25)*rootOf(0.04)) = 0.05
 

LankyLint

Member
Sachal,

Risk spectrum is the general form of both VaR and ES

Risk spectrum is used to assign weights to losses. VaR assigns 100% weight where alpha = 5% (or 1%)

ES assigns equal weights to all losses beyond the VaR.
 

gautamFRM

New Member
I felt exam was not easy....but not tough also...purely based on concepts.
I am not very comfortable in the question which were purely theoretical.....so cant bet on the result. generally how much is the %tile for the part1 ?
 

Hend Abuenein

Active Member
No there wasn't.
And that makes lankylint's answer a bit puzzling for me.

Would you please explain Lanky, why is it that you used 1 for weight of both assets?
 

LankyLint

Member
I heard that it used to be pretty low. 55/100 used to be enough.

I bet it will be higher this time (although we can only speculate).
 

LankyLint

Member
Hend,

I understand your point. However, the weights were not explicitly mentioned. My first instinct was the same as yours and I got some 4. or 5. something answer.

However, since the portfolio variance is given, we don't have to worry about the weights. Whatever they are, they have been incorporated into the calculation for portfolio variance. Using the rest of the information, the correlation would make the variance (and implicit weights) equal to the portfolio variance.
 

Hend Abuenein

Active Member
I don't think that's very accurate, since the weights affect first and second term in the variance formula as well, and that effect cannot be captured by the covariance nor the correlation.

Thanks for sharing opinion
 
Top