Exam Feedback May 2019 Part 2 Exam Feedback

[christylee]

Isnt the answer for maximizing sharpe ratio > increase CAD and decrease the other two? I thought so cuz return/beta and the sharpe ratio was the highest for CAD

 

[jkowalonek]

Hello
I took the Exam in Amsterdam this Saturday.

I studied close to hundreds of hours and undertook the necessary expenses.
This was very enjoyable, if difficult. Certainly no guarantee of passing. But dedicated my best efforts - pretty muich like everyone else I suppose.

On the day, after stressing that I adhered to all the Garp rules - I pass through security and finally arrive at my desk. I was so afraid my phone would go off - even after checking it an imaginary 6 times that I did not even risk bringing it in the end and came with only my door keys, calculator and id.

I sit down to start the exam - and there was some expected noise from staff organizing boxes and papers. I anticipated this would calm down after some minutes so tried to settle into exam mode. 10 mins - 20min - 30mins - 40mins - the whispering goes on, is louder and even in a joint group. Like a Saturday night out.....

Two hours into the exam - I finally have the courage to raise my hand
The guy was plesant enough, said sorry and stopped. A short time later 4 exam attendants were sitting together whispering again. It was not a joke. 30 mins left with adrenaline running and you cannot concentrate..

I wanted to share my experience to see if anyone else had a similar experience. Or just get some feedback

To me, one central tennet of the FRM is about culture and behaviour and I find it hard to believe that Garp could support such behaviour


Thanks
Joya

We had a very similar issue with noise, they changed the venue for the Charlotte, NC exam from the University to a hotel. The hotel had a Wedding going on next door at the same time. The first hour of the exam wasn't bad, but then the noise started to pick up and kids were screaming and crying outside the ballroom. They had the food for the wedding set up in the lobby right outside our "exam room". Toward the end of the exam it got so loud I had to stick my finger in my ears just to read a question . But I digress, if it was a real life situation, I'd have managers breathing down my neck to figure out hedge ratios and VaR calcs

 

[Yogi88]

Does anyone remember the answer of lognormal var question with 414 and 416 options?

 

[kylian.mbappe]

VAR

Does anyone remember the answer of lognormal var question with 414 and 416 options?

yes.. to me, it was a big trap.

The formula of lognormal VAR i.e P*[1 - exp(mu - z*sigma)] gave 387
The formula of normal VAR i.e P*[-mu + z*sigma] gave 416
The option were 414 and 416 and the question was "what is the closest to the lognormal VAR ?"

Thus, I put 414........ Am I the only one ?

 

[nikic]

yes.. to me, it was a big trap.

The formula of lognormal VAR i.e P*[1 - exp(mu - z*sigma)] gave 387
The formula of normal VAR i.e P*[-mu + z*sigma] gave 416
The option were 414 and 416 and the question was "what is the closest to the lognormal VAR ?"

Thus, I put 414........ Am I the only one ?

I must've not had an issue getting the answer from Lognormal VAR as I simply do not recall this question.

Is this the one where they gave 252 trading days (and you had to convert the annual std deviation to daily), or a totally separate question?

 

[kylian.mbappe]

I must've not had an issue getting the answer from Lognormal VAR as I simply do not recall this question.

Is this the one where they gave 252 trading days (and you had to convert the annual std deviation to daily), or a totally separate question?

Yeah this one

 

[Yogi88]

I must've not had an issue getting the answer from Lognormal VAR as I simply do not recall this question.

Is this the one where they gave 252 trading days (and you had to convert the annual std deviation to daily), or a totally separate question?

Understood..made a mistake here....also do you remember the answer for ho-lee model.was it 2.xx?

 

[you5135]

dose anyone remember the question of unsmoothing ?

 

[nikic]

dose anyone remember the question of unsmoothing question?

Yes, the answer would be along the lines of unsmoothed returns has a lower sharpe / higher volatility.

Do you recall the answer choices by any chance?

 

[you5135]

what was the question?

 

[nikic]

what was the question?

I think it was basically asking whether Sharpe ratio and volatility would be higher or lower with if they smoothed the returns on illiquid assets.

 

[amit.m.sharma]

what was the question?

Is this is the one where a firm has new funds coming in which were to get equally allocated between two managers having the same sharpe ratio. Then one manager is upset because the other manager is trading on illiquid assets and has an inflated sharpe ratio? Or is this another question...?

 

[tirerasalim]

yes.. to me, it was a big trap.

The formula of lognormal VAR i.e P*[1 - exp(mu - z*sigma)] gave 387
The formula of normal VAR i.e P*[-mu + z*sigma] gave 416
The option were 414 and 416 and the question was "what is the closest to the lognormal VAR ?"

Thus, I put 414........ Am I the only one ?

I went with z= 1.645 and not 1.65 and found 414 also, closest!

 

[Amarnadh D]

Thank you @nikic really appreciated, were the choices only average of worst 5 or 6? Because I want to contact GARP if they did this incorrectly. If it was 98% with 252 Trading days, it seems unlikely to me they offered the correct solution ....

Hi David, it was 252 days only@ 98% C.l

 

[Amarnadh D]

My understanding to this question is 2% of 252 day is 5.04, ES takes the weighted average beyond the significant level. Therefore start from 5 and then 4,3,2,1 are points beyond the significant level of 5.04 which should be added to take the average.

Agreed. In fact, I've got the same answer..

 

[David Harper CFA FRM]

Agreed. In fact, I've got the same answer..

@Amarnadh D thanks but what same answer? I have a call into GARP (also in regard to the unconditional PD question) about this. If the assumption was 98% and 252 trading days then the correct answer is neither the simple average of worst five or six but rather, per my post above, the correct answer would need to include a slight weight for the sixth worst, as given by a non-simple average. Maybe the solution was correct, I can't tell by the feedback ....

As I wrote above, if the 6 worst losses were: -2.8%, -2.6%, -2.5%, -2.4%, -2.3% and -2.2%
... then the correct answer for the 98.0% ES is 2.5175% (not 2.5200% and not 2.4667% which are simple averages; but maybe the answer only displayed 2 decimals such that 2.52% was given (?), and that would be a happy coincident outcome, although I'd still want to talk to GARP if they aren't doing this correctly and it only happens to be correct b/c the displayed options were limited to, eg, simple averages 2.52 and 2.47). I'm skeptical simply because the setup of 98.0% ES and T = 252 days automatically throws ES into a harder calculation, versus it's much easier to request 98.0% ES and T = 250 days because that is a simple average (assuming unweighted HS) of the worst 5.0 = 250 days * 2.0% losses.

 

[amit.m.sharma]

@Amarnadh D thanks but what same answer? I have a call into GARP (also in regard to the unconditional PD question) about this. If the assumption was 98% and 252 trading days then the correct answer is neither the simple average of worst five or six but rather, per my post above, the correct answer would need to include a slight weight for the sixth worst, as given by a non-simple average. Maybe the solution was correct, I can't tell by the feedback ....

As I wrote above, if the 6 worst losses were: -2.8%, -2.6%, -2.5%, -2.4%, -2.3% and -2.2%
... then the correct answer for the 98.0% ES is 2.5175% (not 2.5200% and not 2.4667% which are simple averages; but maybe the answer only displayed 2 decimals such that 2.52% was given (?), and that would be a happy coincident outcome, although I'd still want to talk to GARP if they aren't doing this correctly and it only happens to be correct b/c the displayed options were limited to, eg, simple averages 2.52 and 2.47). I'm skeptical simply because the setup of 98.0% ES and T = 252 days automatically throws ES into a harder calculation, versus it's much easier to request 98.0% ES and T = 250 days because that is a simple average (assuming unweighted HS) of the worst 5.0 = 250 days * 2.0% losses.

Hi David, I believe the next loss was indeed 2.2% as you assumed and the second option was 2.52%

 

[amit.m.sharma]

Isnt the answer for maximizing sharpe ratio > increase CAD and decrease the other two? I thought so cuz return/beta and the sharpe ratio was the highest for CAD

I chose the same

 

[Bhandaripulkit]

Isnt the answer for maximizing sharpe ratio > increase CAD and decrease the other two? I thought so cuz return/beta and the sharpe ratio was the highest for CAD

Return /beta was same for all the position as far as i remember

 

[amit.m.sharma]

Understood..made a mistake here....also do you remember the answer for ho-lee model.was it 2.xx?

2.08? I remember ‘k’ the rate of mean reversion was too small, 0.02