FRM exam level1

[mgarcia]

I read the comments on about the level 2 exam but did not see more comments on level 1.

The level 1 exam was long and with too many calculations. Having taken level 1 and 2, 80 questions in 4 hrs is doable. You have time to think and understand the poorly written questions. The questions were not hard in general but there were too many calculations, more than 60% of the question where calculations…so few quality questions where you are tested on concepts. I think GARP exam is terrible and it does really test if you know risk management. Being able to punch numbers fast in calculator does not prove you learned or/and understood the concepts. There are very few quality questions testing the concept behind the concept …. Any way, from what I can remember:
- two questions about the GARP rules of conduct. They were terrible written and you had to read them more than once to make sense of what was being asked
- the question provided the data results from a regression and asked what why the results were wrong. I think the answer was that sum (eiXi) was no zero.
- 5-6 questions on put-call party: find the dividend, find the price of the put and then answer if it was under priced or overpriced.
-5 question on commodities: convenience yield,
-3-4 questions about numbers of contracts to buy or sell to cover your risk using duration, correlation, there was one with B (quite difficult, someone there remembers the question)
- there was a graph about the basis risk and correlation and you had to answer how the basis was being affected …someone there remembers the question?
-One question about conditional probability.It was not hard…but again you need to think, and then punch the numbers … it takes time
-One question provided the probabilities for5-7 values and asked you to calculate the tracking error…when I saw the question I laugh…I just jump the question…you had to calculate the mean and the use the value to calculate the variance
- another similar question, gave you the data for 7-8 values, and you needed to calculate the Kurtosis…also jumped the question. Too many calculations with very little time!
-Expected short fall. The question provided a table with data and you had to calculate the value.
-4-5 questions on EWMA and garch. You had to calculate the long run variance, another where you had 3 GARCH models with different persistence and you where asked which one had a slowest regression to the long-run variance.
- two binomial trees with two levels. One you needed to calculate the value of a put. Someone there remembers the question.
Question 99, it was about an importer who needed to buy JPY fwd, the asked for how the payout would look and they provided formulas…one of them was 450,000-500,000/fx…my time was almost up so I just went for intuition. Someone there remembers the question?

Having taken other courses before for FRM preparation, I can say Bionic Turtle videos and applications are the BEST. You boot-camp questions are great. You additional questions to a question truly help grasp the concepts. I will suggest for you to add some FULL practice exam that mimic the exam conditions. Sadly, for Level 1, you need a lot of practice using your calculator, you need to be very fast and accurate…there is no time for reentering!
Thank you!

 

[David Harper CFA FRM]

Hi MGT,

Thank you for a thoughtful, detailed update. It is really helpful; e.g., a lot of Hull, volatility.

Good suggestion re: full practice exam. I think we will try to add this (and flashcards per the other suggestion). I have noted specifically to discuss, with the team, how we can do this.

I hope you get a chance to relax a bit, I'd love to hear when you get your result (I am optimistic in your case especially). Thank you for your forum participation this season!

David

 

[emer]

Here my feedback,

not happy at all. A lot of weird calculations which were pretty time consuming. Do not know, have been consitently scoring 75 - 85 % in practice exams and needed much less time. Now I am convinced that I will become a returning candidate :-(

[...]
- the question provided the data results from a regression and asked what why the results were wrong. I think the answer was that sum (eiXi) was no zero.
[...]

sum (e_i) was not zero either (around 17,000 if I recall)==> my choice. But thought to chosing
sum (e_i*X_i) either.

Some other questions I remember

(1) Calculating number of years for alpha being stgatistically significant at 95%. I chose 17 years, but this is wrong since I used 95-quantile instead of 97.5% (2-sided-test)

(2) Calculating number of futures to short to reduce portfolio duration to 0.5 fpr a portfolio of size 4.2 Mio with a beta of 2.2, future price of 9665, contract size 5 ==> answer should be 141

(3) 1 question involving calculation probabilities from a binomial distribution: 10 bonds given with equal probability of default. What is the prob. of having 1 or 2 defaults in a year.

(4) Given 3 assets all with equal variance and covariances all equal to 0.5 ==> calculate the minimum-variance portfolios variance ==> for me 2/3 (equal weight for all assets?)

(5) Using GARCH to update a correlation coefficient estimate. However before you had to calculate the returns for the 2 assets, then update the volatiliy estimates, claculate the previous covariance estimate from the previous correlation coefficient (given) then updating the covariance estimate and finally putting all together into the new covariance. Sound straightforward, but I did not manage to do it, all my results did not match one of the answers.

(6) A regression returns stock X on stock Y; intercept and slope coefficient not given. Betas given with regards to an Index for both stocks and risk-free rate. Question: What would the intercept most probably look like in the CAPM? I chose the risk-free rate but not sure if it is correct, I doubt it.

(7) Calculate change of an option value if volatility changes from 20% to 40% (no vega given) using a binomial tree. Value of at-the-money call given (15.5) and current stock price (not sure, think 100)==> My result 38%; Calculations involved setting u, d according to CRR, from that calculating the risk-neutral probability then valuing the option for new volatility. Taking the difference and setting this in relation to the original option value.

(8) J-B-Test: p-values given (very,very small). Question (combined answer): Will H0 be rejected and what is the asymptotic distribution of the test statistic: My answer: chi2 & reject H0.


That's it for now,

Regards

 

[mcarthur724]

I took Part 1 exam in Denver. There were 8 people signed up for the exam here, but only 3 out of 8 showed up, so we had a tiny group of just 3 taking it I had a similar opinion of the exam as the others have expressed here. My biggest complaint is that so many of the questions were able to be narrowed down to 2 choices, but then there was so much ambiguity in the wording that you couldn’t determine which was the appropriate answer. This was especially true of the qualitative questions. Below are some of the additional questions that I remember. The first couple are examples of questions on very basic topics that I knew very well, but could not determine the appropriate answer due to the ambiguity.

1) Question about when it is optimal to exercise early. Of the two choices that I was able to narrow it down to were A) It is not optimal to exercise a European call option early. D) It is optimal to exercise an American put option early when it is just in the money. Regarding “D”, I know that the rule is that American put option may be optimally exercised early if they are significantly in the money, so the use of the word “just” in this question made me think that they were referring to “just = insignificantly in the money” and so this answer is probably not right. However, “A” was also worded strange to say that “it is not optimal for a European call” to be exercised early because European cannot be exercised early. So I went with “A”, but didn’t feel confident because to say that it is not “optimal” to exercise European early is a strange way to phrase it when it should truly be worded as “cannot”.
2) Code of Conduct question – It said that an FRM holder told a committee that “VAR represents the maximum potential loss” to the company. First of all, it didn’t specify that this committee was internal of her company, but that is what I had to infer from it. The two potential choices here were A) She violated the code of conduct in this statement to the committee. B) She did not violate the code of conduct. This was difficult for me to interpret because it wasn’t clear whether this was just a matter of her mis-speaking to a committee at her firm, or if it was a matter of her purposely mis-stating the definition of VAR to try to understate the true risk exposure of the firm. Due to the ambiguity it was difficult, but I went with the first choice of a violation of the code assuming that they meant that she pursposely mis-stated VAR to the committee.
3) Question about which of the following distributions is exponential: A) Wiebul B) Pareto C) Poisson D) Gamma This was one of the rare examples of a straightforward question.
4) There was a question that seemed very straightforward about which of the following forward prices presented an arbitrage opportunity. It gave the commodity’s risk free rate, lease rate, and convenience rate. It seemed totally straightforward and simple, but I just couldn’t come up with the correct answer from the choices. I spent quite a while on this one trying to figure out what trick there was that was making me not get one of the choices, but couldn’t figure it out.
5) A question involving the basic credit rating migration matrix. I had done this many times easily in practice questions but somehow couldn’t get the answer this time. Looking back, I think that it gave a starting credit rating of AA and I accidentally started with AAA. Of course, I never had time to go back and try to check what mistake I had made since I ran out of time. I ran out of time with about 4 or 5 questions remaining.

Thanks again David for your great service.

 

[Quest4FRM]

I took the Part 1 exam in Singapore...we had quite a few people here for the FRM exam...I agree the test was very quants oriented...lots of calculations...felt short of time in the end..and had to randomly guess a few answers...i also agree with cpaguy7 to the extent that a few questions were ambiguously written...but I guess that was just a trap set up by GARP...

"1) Question about when it is optimal to exercise early. Of the two choices that I was able to narrow it down to were A) It is not optimal to exercise a European call option early. D) It is optimal to exercise an American put option early when it is just in the money."

Dnt remember wat the other 2 choices were, but I think the answer was one of the other 2 only...since I ruled out both of these choices mentioned above due to the ambiguity mentioned by cpaguy7...and i think I still found answer for this question from the leftover choices (if i remember correctly something to do with interest rates and option value)....not sure though now after seeing this post

"2) Code of Conduct question – It said that an FRM holder told a committee that “VAR represents the maximum potential loss” to the company. First of all, it didn’t specify that this committee was internal of her company, but that is what I had to infer from it. The two potential choices here were A) She violated the code of conduct in this statement to the committee. B) She did not violate the code of conduct. "

I think this shud be violated the code...since she is mis-stating the VAR by defining it as maximum loss, this might be her opinion but certainly not the fact, since VAR is not the maximum loss...

"4) There was a question that seemed very straightforward about which of the following forward prices presented an arbitrage opportunity. It gave the commodity’s risk free rate, lease rate, and convenience rate. It seemed totally straightforward and simple, but I just couldn’t come up with the correct answer from the choices. I spent quite a while on this one trying to figure out what trick there was that was making me not get one of the choices, but couldn’t figure it out."

Even i didn't get any of the choices...but the question was "which of the following forward prices presented an arbitrage opportunity"...there was 1 option with 3 out of the 4 choices mentioned...i thot that correct forward price could only be 1 out of those 4 choices...so out of 4, 3 choices should present arbitrage opportunities (either go short or go long)...so i picked the choice with these 3 choices ... the assumption was that out of all data given in the question we could get only 1 forward price out of those 4 choices...so other 3 choices have to be arbitrage opportunities...i knw this was a big and most probably a wrong assumption

"- the question provided the data results from a regression and asked what why the results were wrong. I think the answer was that sum (eiXi) was no zero."

For this, i marked sum(ei) was not zero, as the answer....i thot sum(ei) shud always be zero and that is precisely why we take sum of square of ei to negate the positive and negative values, dnt knw how true i am there

"- there was a graph about the basis risk and correlation and you had to answer how the basis was being affected …someone there remembers the question?"

I think this was that the variances of spot and futures was 1...and we had to find graph representing relationship between variance of hedge and correlation coefficient...i think i picked up one of the straight line graphs as the answer...since the values were satifying the equation that i derived...

"-One question provided the probabilities for5-7 values and asked you to calculate the tracking error…when I saw the question I laugh…I just jump the question…you had to calculate the mean and the use the value to calculate the variance"

Not exactly sure wat the question was, but i think there was a simpler way of doing that...not entirely sure but...need to remember the exact question to figure out...

"- another similar question, gave you the data for 7-8 values, and you needed to calculate the Kurtosis…also jumped the question. Too many calculations with very little time!"

This is the excess kurtosis one i think...here also i guess u didnt need to calculate the kurtosis...i think the distribution was normal...so excess kurtosis was 0...

"Question 99, it was about an importer who needed to buy JPY fwd, the asked for how the payout would look and they provided formulas…one of them was 450,000-500,000/fx…my time was almost up so I just went for intuition. Someone there remembers the question?"

I too marked this option...was short of time to calculate this...and 2 options were payoffs for options...like max(0, some value)..i thot since the question is talking abt forward contract, the payoff can't be 0...it needs to be some value, either positive or negative...so I picked this choice as the answer...

 

[Quest4FRM]

Some of the other questions I can remember:

1.) derivatives dealing between P&G and Bankers Trust...which GARP code of conduct was violated by Bankers Trust?
I marked the last option, dnt quite remember what it was....but at that time it appeared to be the most logical one to me...

2.) Dnt remember exactly...but 1 question on bank has reserved economic capital based on VAR and taken insurance for other risks...what happens to risk...choices were like: (a) correlation risk increases, credit risk decreases (b) market risk decreases, credit risk increases ...something of this sort ...does someone remember the exact question?

3.) Compared to corporate default rate, sovereign default rate is : (a) significantly higher (b) higher (c) lower (d) some other option ... was sure it was between (a) and (b), and finally marked (b)

4.) Find the CTD bond...all values were given..just needed to knw the formula

5.) which distributions can be used to model frequency and severity of operational risk? The answer was Poisson and Lognormal

6.) One good question was: VAR calculated at 95% and 99% based on "full revaluation" and "delta normal" approach...it asked 2 things...there were 2 blanks..dnt quite remember exactly..but i was confused on this one...

7.) question on which of these on its own is not a significant reason for risk mgmt
(a) Variable cash flows
(b) Tax rate
(c) Debt overhang
(d) Financial distress
was confused between (a) and (b)...and marked (a) i guess...i think i m wrong on this one...dnt knw what the right answer is...

8.) compared to coupon bonds, zero coupon bonds have the lowest <blank> ... answer was reinvestment risk

9.) one graph was given and regression equation was asked based on 4 choices...was easy...

10.) There was 1 good question...went something like...4 traders and their positions were given...trader A has a long position in 1000 calls at strike price 80...tarder B has a long position in 500 calls at strike price 100 and short position in 1000 puts at strike price 100 blah blah...dnt remember what the question asked..but this was an indirect question on box spread...and i marked the answer accordingly

11.) Lots of questions on bond calculations...spot rates, forward rates and stuff...

12.) when is credit rating agency's ratings not transparent <this is very badly worded by me...this wasnt the exact question, but it meant the same > ..the answer was issuer pay model

13.) advantage of top-down approach of operational risk over bottom-up approach...
(a) can allocate capital business unit wise
(b) can allocate capital on a firm wide basis..and some other choices...
dnt remember what I marked...was confused between these 2...anyone knows the answer or other options??

 

[svnitparas]

Some more questions:

1) You were given Variance(A) and Variance(B) and corellation, need to calc Variance(A+B).
2) One qs each on Expected and Unexpected Loss calculation.
3) ES calculation for VAR(95.5 - 99). I Have taken Avg of all given values.
4) Debentures have claim above what all .. options were MBS , ABS , Subordinated Debentures , .... I marked Sub-Debentures only.
5) One qs for margin call... Dont remeber exactly but I marked the option with 2 days I think July 2 and July 3 Only.
6) One qs about treynor , Jensen , Sharpe...I marked 4th answer which was Treynor uses systematic risk while sharpe uses total risk.
7) One straight qs on calculating Hedge ratio given Corellation , variance of spot and variance of futures.

 

[Quest4FRM]

Some more questions:

1) You were given Variance(A) and Variance(B) and corellation, need to calc Variance(A+B).
2) One qs each on Expected and Unexpected Loss calculation.
3) ES calculation for VAR(95.5 - 99). I Have taken Avg of all given values.
4) Debentures have claim above what all .. options were MBS , ABS , Subordinated Debentures , .... I marked Sub-Debentures only.
5) One qs for margin call... Dont remeber exactly but I marked the option with 2 days I think July 2 and July 3 Only.
6) One qs about treynor , Jensen , Sharpe...I marked 4th answer which was Treynor uses systematic risk while sharpe uses total risk.
7) One straight qs on calculating Hedge ratio given Corellation , variance of spot and variance of futures.​

Yea i also marked the same answers for all these questions..cheers!!!

 

[braxxus]

I don't even remember some of those questions, I have tried to forget the experience at this point in time. I agree that the ethics questions were some of the worst written questions I have seen in my life. I can't remember more than a handful of questions, but even then I don't remember the wording or anything.

I looked at the Schweser 2009 FRM notes in addition to BT. The BT information was better other than I hated to read PDF files. If I had only looked at the Schweser stuff, it would have been a complete slaughter. I guess we will all see how we do in January. It depends on how well the top 5% did or so as they supposedly create the curve. I felt that I made educated guesses on probably 45-50 questions and flat out guessed on another 5-10.

I didn't feel confident coming out of either CFA Level II or III, but I passed both of them. I feel less confident in this exam, but I guess a lot of other people thought the exam was too much. Curious how many people took the exam at your site? I was in Philadelphia and only about 15-18 people sat for the exam, which is much smaller than the number that sat for the CFA at a much smaller location when I took that exam.

 

[mcarthur724]

3.) Compared to corporate default rate, sovereign default rate is : (a) significantly higher (b) higher (c) lower (d) some other option ... was sure it was between (a) and (b), and finally marked (b)
The answer that I chose for this question was (d) similar. I think I remember seeing a comparison between Corp and Sovereign default rates that showed them to be similar.

There was also a case study question about lessons learned from Kidder-Peabody. The 2 choices that I had narrowed it down to were (a) separation of the back office function (b) management should be aware of the source of unexpected large dollar profits. I chose (b).

 

[Quest4FRM]

In Singapore there were many people..both for FRM as well as ERP exam..i think both between 60-70...FRM maybe a bit higher...
I think number of ppl giving FRM is bound to be lesser than CFA, since FRM is very specific to risk professionals only while CFA is much broader and hence almost everyine who is into any finance related job(risk or otherwise) gives it

 

[Quest4FRM]

3.) Compared to corporate default rate, sovereign default rate is : (a) significantly higher (b) higher (c) lower (d) some other option ... was sure it was between (a) and (b), and finally marked (b)
The answer that I chose for this question was (d) similar. I think I remember seeing a comparison between Corp and Sovereign default rates that showed them to be similar.

There was also a case study question about lessons learned from Kidder-Peabody. The 2 choices that I had narrowed it down to were (a) separation of the back office function (b) management should be aware of the source of unexpected large dollar profits. I chose (b).​

Yea i too narrowed it down to these 2 options but marked (a)
Frankly speaking i hav no idea what the answer is...

For the sovereign default rate i still think it is higher...my logic was that there are only less than 200 countries...even if 2 default the default rate wud be 1%...but there r tens of thousands of corporates...to achieve the same 1% default rate, a significantnumber of corporates wudneed to go into default....
Wishful thinking i guess

 

[girishkhare]

Compared to corporate default rate, sovereign default rate is : (a) significantly higher (b) higher (c) lower (d) some other option ...​

Governments have extra arm of defense against default that corporates don't have i.e. governments can print more money and meet their bond coupon or principal obligations. Corporates cannot do that. Of course in the process, inflation increases but technically, government can meet its obligations by printing more money. Given this, shouldn't the answer be (c) or lower than corporate defaults?

I used this logic and marked (c) in FRM.

 

[mcarthur724]

I am almost positive that there was actually a specific table in the Schwesser notes that pointed out that Corp & Sovereign default rates are similar across each credit rating. Therefore, I’m pretty sure that the answer to this one was (d) similar.

 

[Quest4FRM]

Well technically we all can be right and wrong as well...i m sure at higher ratings the default rate of corporates and sovereign wud be similar or less infact for sovereign...but at lower ratings sovereign default wud b higher...dunno wat GARP was looking for

 

[svnitparas]

One more qs out of my memory in which Comparitive advantage was given in a swap and in which the mediator is offering 10 basis point advnt to both the parties. Need to calculate how much is the mediator basis profit?

I marked answer as 30 basis point since comparative advntg was 40 basis subtract 10 basis = 30 basis…what was your answer?

 

[mgarcia]

One more qs out of my memory in which Comparitive advantage was given in a swap and in which the mediator is offering 10 basis point advnt to both the parties. Need to calculate how much is the mediator basis profit?

I marked answer as 20 basis point since i subtract 10 basis for each side

 

[mcarthur724]

I selected "20 bps" for that question also, rather than 30.

 

[svnitparas]

One more goes down as incorrect for me. God save me.

 

[braxxus]

I selected "20 bps" for that question also, rather than 30.​

Yeah, definitely 20 bps as i recall that question.