PDA

View Full Version : Ridiculous


smekta
05-04-2011, 01:59 PM
Had 6-7 questions completely unanswered on the full 5 when they called time and Im usually one of the fast ones, anyone else in the same boat?

herr blau
05-04-2011, 02:08 PM
+1

And I'm mf'n p'd about it

Adversely Selected
05-04-2011, 02:17 PM
Yup, I had about the same. I am pissed, maybe everyone will be in the same boat.

tommie frazier
05-04-2011, 02:24 PM
too soon for discussion?

monkey2003
05-04-2011, 02:27 PM
how many freaking simple averages and link ratios do they want on the exam. i mean cmon they take lots of time. given that i only got through 60 points, and left the other 20 points blank. not because i didn't know it, but because i ran out of time. they passing grade better be below 70%!

periculum
05-04-2011, 02:28 PM
I took 5A and I lost the last question because I ran out of time. Basically lost 8% of the points right there.

I knew this would happen.

actuarialista
05-04-2011, 02:44 PM
Agree totally. I managed to finish the practice exams in the time allowed, but today (on full 5), could not get it done. Left at least 8 points unanswered out of the 79.5 or whatever it was.

Bobby
05-04-2011, 02:46 PM
I had about 4 points that went unanswered. Another 4 points were very quickly answered but should get most credit for it. Felt like the material wasn't too difficult, but it was just a long exam.

smekta
05-04-2011, 02:46 PM
how many freaking simple averages and link ratios do they want on the exam. i mean cmon they take lots of time. given that i only got through 60 points, and left the other 20 points blank. not because i didn't know it, but because i ran out of time. they passing grade better be below 70%!

I guess for next time Ill just practice my calculator number crunching skillz and ignore everything else since I wont get to it anyway. uber retarded.

fsumath
05-04-2011, 02:50 PM
Yeah the exam was too ridiculous. I actually didn't finish the last 5 or 6 questions. I also skipped a couple of the ratemaking questions because I couldn't remember how to do them. Also I felt some of the reserving questions were not consistent with the syllabus.

kyleucf
05-04-2011, 02:50 PM
Full 5, had about 5-6 I left blank (but at least two 3 pointers). I did reserves first, didnt think it was too tough til i looked at the clock, took me 2:20-2:30 so not alot of time left for rates. Rates was tougher than reserves imo... Def would have passed 5B, not 5A.

fsumath
05-04-2011, 02:53 PM
Okay at least I don't feel alone this time. This exam should be thrown out. I know it won't be but it just was so unfair. I made my points on CAS survery.

snoo
05-04-2011, 02:56 PM
I am sorry to hear about the ratemaking piece. I was able to finish 5b in the time allotted. But i only had to study for 5b.

MoFBMel
05-04-2011, 02:59 PM
I was hesitant to check the AO since I usually feel more discouraged after doing so, but I'm glad that I'm not the only one who felt that this exam was long. I'm usually one to finish an exam with time to spare, but I wrote up to the very last second leaving 4 or 5 complete questions only partially answered (I tried to write a little something for everything). Hopefully, this will result in a lower passmark (since it sounds like a few people had a hard time finishing).

HasilAtkins
05-04-2011, 03:04 PM
I am not sure that 5A was harder than 5B or vice versa. I ran out of time on the reserving part, but that is because I spent 2.5 hours or so on ratemaking, not because I found reserving difficult. In fact, I didn't run into a question that I could not answer, but I did have to leave 6-7 blank. I can't imagine that it will be enough for a pass.

I do feel like they made each half longer to really give the half exam sitters a good workout, but that made the full exam very long. I will echo what others in this thread are saying - it is annoying to that feel that you know the material well but to be defeated by the clock.

JOLE5646
05-04-2011, 03:05 PM
how many freaking simple averages and link ratios do they want on the exam. i mean cmon they take lots of time. given that i only got through 60 points, and left the other 20 points blank. not because i didn't know it, but because i ran out of time. they passing grade better be below 70%!

just do it straight up on your calculator...no need to write down the number for each year and take the average. Shouldn't take any longer than the volume weighted avg.

Anyway, I agree with all of you that it's a long exam. I finished 20 min early but only because I was blazing through the questions. I also took the 15 min to mentally do 3 longest questions in my head...so already I had a head start.

I can't believe they didn't ask any question about the complement or CH14 of W&M.

fsumath
05-04-2011, 03:19 PM
Yeah they didn't ask about the complement on last year's exam too. Getting tired of taking this exam over and over again. Looks like it will be my 3rd attempt next year.

folasona
05-04-2011, 03:36 PM
- it is annoying to that feel that you know the material well but to be defeated by the clock.

Unfortunately the number of POINTS some ppl are complaining about missing out on is the number of PROBLEMS I didn't even get to start.

To top it all off, I'm riding the PATH train home and the TV screens decide to flash a PATH to perseverance quote: "It's not that I'm so smart, I just stay with problems longer than other people" -Albert Einstein.

Dunno if that makes me feel better or worse, I really wasn't trying to stay with the problem of how to pass an exam that tests button pressing skills as much or more than thinking skills for whole 'nother year, but whatever...

On the other hand, there's still a tiny sliver of hope, because people who say "I'm usually one of the fast ones" are complaining about the length, but time has always been a killer factor for me on every actuarial exam.

snoo
05-04-2011, 03:38 PM
I did barely get through the read through on 5b. Cant imagine what it was like for full 5.

kyleucf
05-04-2011, 03:41 PM
the read through was the biggest advantage to taking half.

snoo
05-04-2011, 03:48 PM
I am a fast one and had about 7 min left over on 5b. I made/almost made some ridiculous mistakes on 5b.

Vorian Atreides
05-04-2011, 03:58 PM
I could've used an extra 30 minutes . . . would've been happy with even an extra 15.

Agree with kyle . . . reading time was an advantage . . . I was able to work through the answers for the problem I first wrote out.

Present Value
05-04-2011, 04:11 PM
welcome to upper levels...you are not fast if you haven't demonstrated speed by successfully completing atleast a few of the upper level exams under time pressure...i can sympathize with the feelings people have here, but that's what these are about.

Number Theory
05-04-2011, 04:12 PM
Not sure that I could have finished that test if I had been copying directly from the answer key.

monkey2003
05-04-2011, 04:17 PM
Not sure that I could have finished that test if I had been copying directly from the answer key.

hehehe. I completely agree. This exam was by no means hard, but it was so long that no one finished. This is my first exam that i took that is solely administered by the CAS. Not impressed.. Not impressed at all. First sending me the wrong candidate number, and not this? sorry, i'm a bit frustrated today.

Need Vacation
05-04-2011, 04:18 PM
Not sure that I could have finished that test if I had been copying directly from the answer key.
:iatp:

chaoren
05-04-2011, 04:31 PM
I left maybe 10 points blank, most of them are from Ratemaking , but I finished that 6.75 mega one. IMO, that question was ok

BTW I started from Reserving questions first, which I feel I can gain some confidence at first.

chaoren
05-04-2011, 04:31 PM
repeated =x

oblivious
05-04-2011, 04:36 PM
welcome to upper levels...you are not fast if you haven't demonstrated speed by successfully completing atleast a few of the upper level exams under time pressure...i can sympathize with the feelings people have here, but that's what these are about.

I have passed an upper, and think that all of the problems would have been tough to complete in the allotted time at least for 5a., but yes speed is a big part of the uppers.

chopsuey
05-04-2011, 04:38 PM
I got burned by 5A. I'm usually a fast finisher and this was my 2nd attempt of ratemaking. Left 3 blank for time.

Staky41
05-04-2011, 04:49 PM
I also struggled with the time... if the pass mark is 73%, I don't think I have a hope at passing... I left WAY too many completely blank :( So frustrated. I felt like I had the material down. I felt like I would have had to have those questions memorized & show barely any work to finish it.

Galois
05-04-2011, 04:52 PM
I thought the questions were easy, but the test was really really long.

monkey2003
05-04-2011, 04:54 PM
I also struggled with the time... if the pass mark is 73%, I don't think I have a hope at passing... I left WAY too many completely blank :( So frustrated. I felt like I had the material down. I felt like I would have had to have those questions memorized & show barely any work to finish it.

if the pass mark is 73%, my personal guess would be that less than 25% of people would pass the Full Exam 5. From reading the forums, you're one of the lucky few if you managed to finish in time.

earthsunmoonsky
05-04-2011, 04:55 PM
Is the reason most people thought full 5 was so long is because:

1) so many were first time upper exam takers and/or
2) most were first time exam 5 (or old 6) takers?

I thought 5b was long, but doable in the time alloted when writing fast.

However, I can sympathize with those that said 5 was too long as when I failed 6 last time I was in the same boat. I think speed was something that failure taught me more than anything.

Staky41
05-04-2011, 04:57 PM
Does anyone have the link to the survey?? I left it at the exam site...

Second question -- I'm worried I started to really lack on showing my work. Is there a site that explains the grading for what they look for? As in, if I get the right answer and I didn't show work, what do I lose?

Thanks!

Staky41
05-04-2011, 05:00 PM
Is the reason most people thought full 5 was so long is because:

1) so many were first time upper exam takers and/or
2) most were first time exam 5 (or old 6) takers?

I thought 5b was long, but doable in the time alloted when writing fast.

However, I can sympathize with those that said 5 was too long as when I failed 6 last time I was in the same boat. I think speed was something that failure taught me more than anything.


I will admit - it was my first time with exam 5...and my first upper exam. I was frustrated when I walked out... because I am always slow when it comes to exams - usually don't finish them. So I figured it was just me, but surprised to see that lots of people on this forum struggled with the time constraint. At least that gives me hope (since others in my shoes) that I can get faster & do better next time or that that exam was just ridiculously long.

tommie frazier
05-04-2011, 05:03 PM
Does anyone have the link to the survey?? I left it at the exam site...

Second question -- I'm worried I started to really lack on showing my work. Is there a site that explains the grading for what they look for? As in, if I get the right answer and I didn't show work, what do I lose?

Thanks!

there are articles in future fellows, a cas [publication that go over how much is required in describe, briefly describe, show all work, etc.

sassafras
05-04-2011, 05:08 PM
I just don't understand why speed has to be such a big part of the process. Why not make the test a reasonable length and ask questions that require thought rather than fast fingers on a calculator? I guess maybe they would be too difficult to grade....

ReserveRage
05-04-2011, 05:12 PM
I just don't understand why speed has to be such a big part of the process. Why not make the test a reasonable length and ask questions that require thought rather than fast fingers on a calculator? I guess maybe they would be too difficult to grade....

Good idea in theory, harder to implement in practice. The more cute and creative the questions get, the more backlash you see from students claiming the problem is defective, poorly worded, ambiguous, etc.

Often the CAS will take the student's side and toss out part or all of the question.

Galois
05-04-2011, 05:17 PM
I took 6 twice before, and 5 once, and received a 5 on all three attempts, but I still thought it was long. But then I usually do. Time is usually a bigger factor for me that my understanding of the material.

sassafras
05-04-2011, 05:18 PM
Good idea in theory, harder to implement in practice. The more cute and creative the questions get, the more backlash you see from students claiming the problem is defective, poorly worded, ambiguous, etc.

Often the CAS will take the student's side and toss out part or all of the question.

True - But it seems that rather than doing several problems where 90% of the work is taking ratios, they could have tested more that would determine whether you understand what's going on. I took only the reserving part, but it seem very heavy on the menial computation and not so much on the understanding. (But this is me talking after a couple of glasses of wine, so who knows...)

smekta
05-04-2011, 05:27 PM
With 15 minutes left I started doing 1-year volume weighted average for all of my selections, so see ya next sitting :-D

ill go drown out my sorrows :beer:

chaoren
05-04-2011, 05:48 PM
With 15 minutes left I started doing 1-year volume weighted average for all of my selections, so see ya next sitting :-D

ill go drown out my sorrows :beer:

I would give you partial marks if I were the marker !

Present Value
05-04-2011, 05:53 PM
In all honesty, I thought the exam was pretty easy and reasonable

monkey2003
05-04-2011, 06:14 PM
In all honesty, I thought the exam was pretty easy and reasonable

Maybe you should make a thread about how easy the exam was. I sure as hell won't be in there. It was doable, but not in the time given.

Kentucky
05-04-2011, 06:14 PM
Present Value, aren't you sitting Exam 9?

strand
05-04-2011, 06:23 PM
In my opinion, the exam questions were easy, but the exam was too long. I actually would have preferred a shorter exam with slightly harder questions. I felt like I was really prepared for this exam, and so was frustrated that I couldn't answer questions I knew how to solve. Such extreme time constraints don't exist in the "real" actuarial world, and even if they did, we have access to spreadsheet programs that can easily perform the simple, but time-consuming, calculations of age-to-age factors, averages, etc. This exam should have been intended to indicate which candidates have mastered the material. Instead, it will show which candidates can write the quickest and have the nimblest calculator-button-pressing fingers.

periculum
05-04-2011, 06:26 PM
In all honesty, I thought the exam was pretty easy and reasonable

Here come the trolls.:lol:

Bobby
05-04-2011, 06:26 PM
I think they should've changed all "straight average" instructions to "weighted average". Not sure why they feel the need to make us build a whole triangle when each factor is about the same anyway.

Present Value
05-04-2011, 06:29 PM
Present Value, aren't you sitting Exam 9?

yes

Bobby
05-04-2011, 06:36 PM
Exam was surprisingly straightforward. There was a shocking number of questions that seemed to be just lifted from past exams without any modification. Just a couple questions too long.

smekta
05-04-2011, 06:45 PM
Exam was surprisingly straightforward. There was a shocking number of questions that seemed to be just lifted from past exams without any modification. Just a couple questions too long.

Agree, Id give 2 out of 5 for difficulty, but even if it was just a bunch of 2nd grade math problems, I can only do so many in 4 hours.

kyleucf
05-04-2011, 06:48 PM
double sided question sucks...

nonactuarialactuary
05-04-2011, 06:51 PM
I took 5B, but I was the only one in my testing center writing the reserving portion. Everyone else was writing 5A, and their general impression was that it was way too long. I found it interesting that they had 20 questions, while we only had 17 - what was the point breakdown between 5A and 5B (I think 5B had 39 points total)?

I wrote an answer down for 16 of the 17 questions, and only left one question completely blank. I had a few of the subquestions blank too. I failed 6 last fall precisely due to speed. The first few questions on 6 last fall were Berquist-Sherman and Frequency-Severity problems that I wrote complete answers for, which meant that I left a whole lot of points completely blank. This time around, I was a lot more focused on speed, even if that meant showing less work.

Overall though, I have to agree that the way the CAS tests these exams is in need of an overhaul. If you can show that you know how to calculate one or two link ratios, there's no need to show that you know how to calculate a full triangle's worth with 5 years of data. It's nothing but mindless calculator work that's completely irrelevant in the real world, and it detracts from the more important stuff (e.g., you've shown how to calculate losses using the paid loss development method, but why is this method worthless here?)

I'm glad to see that a lot of people on here had issues with the length. Last fall, I expected to see a lot of complaints after the exam was over, but saw people saying how easy they thought it was instead.

Smartbill
05-04-2011, 06:52 PM
There was a shocking number of questions that seemed to be just lifted from past exams without any modification.

I noticed this too and at the time was like huh? I've seen this before...

Kitty Litter
05-04-2011, 06:58 PM
I would say the exam seemed fair, just not enough time to finish all the problems! Why do they keep doing this?!?! Ugh!!! :furious:

sohlelynol
05-04-2011, 07:17 PM
I took the full 5. I thought 5A was very long but not too difficult. The 5B part took me a lot less time and was easy.

But the 5B part was stupid. It seemed like the calculations were pretty much exactly the same for every problem. It was like a test to see whether I'd get bored if I keep repeating the same work over and over.

skaren9
05-04-2011, 07:30 PM
Exam 5 was the dumbest exam i've ever taken! Time was never an issue for me before this one. The questions were easy and similar to previous exams but were waaaaay too many. I would rather have more time to think about some difficult questions than doing all these stupid calculations. I guess CAS was thinking ok, questions are not hard, so, what should we do to make the exam hard? Ah, let's just have a lot of them; hence, it will be hard.

Number Theory
05-04-2011, 07:39 PM
I was rushing so fast once I got to the reserving part all I had time to do for the part C D E and F (that I think almost every question had) was scribble a few words and move on.
For example (if they had asked Briefly describe two situations in which the technique works well. I said something like "long tailed LOB and new program w/o hist. data") ot sure if that will get me credit...
Also I screwed up the premium trend from date on the indication how many points do you think they take off for that if all else is correct?

Vorian Atreides
05-04-2011, 07:47 PM
Sounds like Ken wasn't too far off with the time expectations for his Quizzes/Sample Exams.

Number Theory
05-04-2011, 07:53 PM
No, actually, I was able to finish the last 2-3 on ratemaking and reserving after increasing my speed. Now I think he was underestimating.

booyah81
05-04-2011, 07:54 PM
Also I screwed up the premium trend from date on the indication how many points do you think they take off for that if all else is correct?

.5 or .75 at most?

Number Theory
05-04-2011, 08:25 PM
ok really the ratemaking part was all a blur. I probably got the exposure base overlap ,fallacy quest., assett share model,the big problem minus 1 for incorrect prem trend, but I got the 2 step trend problem, got the premium discount 1... what are some of the other questions that I am forgetting... How many ratemaking points is that total? Maybe 13ish? Hmm... I need like 10 more ratemaking points from somewhere.

ToBeorNotToBe
05-04-2011, 08:41 PM
I left around 6pt blank. What was total points? 75 or 79?

Adversely Selected
05-04-2011, 08:45 PM
I just don't understand why speed has to be such a big part of the process. Why not make the test a reasonable length and ask questions that require thought rather than fast fingers on a calculator? I guess maybe they would be too difficult to grade....

I guess I find it hard for a society to justify their selection process based on whether you can get 70% in 4hrs instead of 70% in 4.5hrs.

I would rather someone be a credentialled actuary who understands the concepts of reserving and not who could calculate 300 link ratios in 4hrs and take a straight average. So many time consuming and repetitive calculations on this exam that do not fully test mastery of the material. Speed is not typically a factor in the real world, so why is there such an emphasis on speed to determine mastery in the exams? Just another way for those before us who had it easier to keep their inflated salaries?

Maybe I'm just ranting, but I do know that the few fellows at my company don't care if it continues to get harder so they keep their inflated salaries.

Number Theory
05-04-2011, 09:03 PM
How did you guys answer the quota share problem?

Invicta
05-04-2011, 09:05 PM
How did you guys answer the quota share problem?

The first year was XOL and second year was quota share. I based my answer simply because I can find a consistent % ceded out on the second year but not on the first one. So don't know if that's a good enough explanation for the XOL.

earthsunmoonsky
05-04-2011, 09:07 PM
The first year was XOL and second year was quota share. I based my answer simply because I can find a consistent % ceded out on the second year but not on the first one. So don't know if that's a good enough explanation for the XOL.

That's what I put too. Something along the lines of consistent 65% net = 35% QS and the other one was volatile (not a constant percentage) implied XOL.

Bobby
05-04-2011, 09:09 PM
The first year was XOL and second year was quota share. I based my answer simply because I can find a consistent % ceded out on the second year but not on the first one. So don't know if that's a good enough explanation for the XOL.

That's what I put too. Something along the lines of consistent 65% net = 35% QS and the other one was volatile (not a constant percentage) implied XOL.

Yeah that's what I did. Gotta be good enough too. They gave us very little data so I don't see what else they could have expected.

Number Theory
05-04-2011, 09:35 PM
That's what I said too but I didn't have much time so I just said #1=XOL and #2=Quota share 65%. Wasn't there a part B and C to that?

ThePerfectGame
05-04-2011, 09:43 PM
Wow. I took 5B and my first few thoughts leaving the exam were about how the exam questions weren't too difficult, but the volume of calculations, specifically simple averages of link ratios, made the exam ridiculously long. I'm not a fast writer at all (though I was able to pass old 5, so I'm not completely untested), but on an exam like the one I just took, I feel like I'd have to be damn near flawless on the questions that I do completely answer in order to have a shot. Granted, I only left one completely blank (a 2-pointer), but several of my written answers from the last 25 minutes of the exam range from "sloppy" to "perhaps written by a second-grader", both in form and substance. Throw in a mistaken Cape Cod when they asked for Expected Claim that I didn't realize until the last 3 minutes (d'oh!!), an ALAE snafu (damn additive), a tail calculation that I only described because I didn't have time to run the numbers, and probably a few other similarly stupid mistakes and I left feeling kinda crappy about it. Hard to believe how confident I felt during the reading period, since it seemed like it was going to be a completely doable exam.

Anyway, at least it's nice to read that others had a similar experience. If the pass score is in the 72-73% range I think my chances are quite slim, but if it dips down to the mid-60s I might have a better chance. Either way, though, I guess there's nothing more I can do about it now, except begin the long slow 8-week death march to the results...

FoxNews
05-04-2011, 09:52 PM
FoxNews has confirmed that the trolls will troll and tell you all to study harder.

FoxNews has also confirmed that you should study harder, trolls or no trolls.

Number Theory
05-04-2011, 09:55 PM
I agree there was only one question that I didn't know how to do on the test and I don't even remember what that one was right now. On the read through I was elated! Then I got to the questions and saw that each one had like 5 parts that mostly required a crapload of work. Happy with the reserving piece though maybe I will get all of that correct and make up for the 7 points that I left blank on the ratemaking exam.

Number Theory
05-04-2011, 09:57 PM
Only one though it was like the 2nd or 3rd problem EC method but they didnt give you an EC ratio. How did you get it?

G. Stolyarov II
05-04-2011, 10:05 PM
I took Exam 5B and managed to give (in my opinion) a decent answer for every question, but just barely; it was a mad race against the clock at the end, and I even blistered one of my fingers while writing. The problems were largely reasonable and recognizable, though I did not appreciate the requirement of simple averages. Anyone who has advanced this far up the exam hierarchy would presumably not have been able to do so had he/she not known how to calculate an arithmetic mean. I do not envy those who had to deal with that drudgery and the ratemaking material in one sitting.

Still, I think I am improving in my ability to complete these exams in the time provided. As with Exam 7 this sitting, I was able to answer all the questions, and I am more confident in my answers here than on Exam 7. An extra 15-30 minutes would have allowed me to provide excellent answers, but the ones I gave might do (I hope).

I agree entirely with those who question whether these exams appropriately test the skills needed to be a successful actuary. If anybody tried to churn out a real actuarial product at the rate the CAS expects us to solve exam problems, I would seriously question the quality. Are we supposed to be thoughtful, critically thinking professionals or frenzied button-pushing trained monkeys whose jobs can be outsourced to Microsoft Excel?

Sincerely,
G. Stolyarov II, ARe, AIS

nonactuarialactuary
05-04-2011, 10:09 PM
Only one though it was like the 2nd or 3rd problem EC method but they didnt give you an EC ratio. How did you get it?

There was one problem where you were asked the ultimate claims for AY2010 or something using the expected claims technique. They gave you premiums, onlevel factors, reported losses, LDFs, and a trend rate. If this is the problem you're talking about, I got the ECR as the all-year average of trended ultimate losses divided by onlevel earned premium.

nonactuarialactuary
05-04-2011, 10:14 PM
That's what I said too but I didn't have much time so I just said #1=XOL and #2=Quota share 65%. Wasn't there a part B and C to that?

No part C - just A and B. Part A was as you describe (I got the same), and Part B asked something about gross and net tail factors I think. It helps that I knew all about reinsurance from sitting for 6 last fall. The problem was tiny, but does Friedland discuss this in enough detail for a numerical problem? If any question was questionable as to whether it's covered on the syllabus, it was probably this one.

Number Theory
05-04-2011, 10:19 PM
jts what did you put for the answer to that one? I think that I put estimate using gross loss not net loss then remove XOL after the fact.

Number Theory
05-04-2011, 10:20 PM
There was one problem where you were asked the ultimate claims for AY2010 or something using the expected claims technique. They gave you premiums, onlevel factors, reported losses, LDFs, and a trend rate. If this is the problem you're talking about, I got the ECR as the all-year average of trended ultimate losses divided by onlevel earned premium.

That's what I did just curious. Then I almost applied the BF method but saved myself somehow.

ThePerfectGame
05-04-2011, 10:20 PM
"No part C - just A and B. Part A was as you describe (I got the same), and Part B asked something about gross and net tail factors I think. It helps that I knew all about reinsurance from sitting for 6 last fall. The problem was tiny, but does Friedland discuss this in enough detail for a numerical problem? If any question was questionable as to whether it's covered on the syllabus, it was probably this one."

FWIW, that's the only one I completely skipped. Didn't recognize it at first blush and didn't get a chance to give it a second look. I'd love it if that one got tossed.

nonactuarialactuary
05-04-2011, 10:39 PM
jts what did you put for the answer to that one? I think that I put estimate using gross loss not net loss then remove XOL after the fact.

Can't remember the exact question, but if they were asking what you would consider for the tail factors, with a quota share arrangement, tail factors for net should be equal to tail factors for gross. For XOL coverage, gross losses can develop more than net losses (b/c they are not capped), so at any given valuation point, gross LDFs would be larger. I think I put something down like this, but I can't remember exactly what they were asking for.

Michael D Wang
05-05-2011, 03:35 AM
My First Upper level exam & first exam with CAS, if previously joint exams do not count.
Guess the one made claims liability part must be a big fan of "being simple" or "being straight". For God's sake, how many times actuaries in practice compute ATA factors manually? Are we still living in 1980s when computers & software were still as remote as China's democracy?
Believe the exam should be more focusing on how to appropriately select ATA factors based on insurer's internal operation & external environment, instead of encouraging "super smart" individuals who can do easy math in their mind faster than calculators. If the latter was the case, why wouldn't they come to China and worship those kids who can be calculus experts at the age of 12.

sohlelynol
05-05-2011, 03:56 AM
My First Upper level exam & first exam with CAS, if previously joint exams do not count.
Guess the one made claims liability part must be a big fan of "being simple" or "being straight". For God's sake, how many times actuaries in practice compute ATA factors manually? Are we still living in 1980s when computers & software were still as remote as China's democracy?
Believe the exam should be more focusing on how to appropriately select ATA factors based on insurer's internal operation & external environment, instead of encouraging "super smart" individuals who can do easy math in their mind faster than calculators. If the latter was the case, why wouldn't they come to China and worship those kids who can be calculus experts at the age of 12.

What the **** are you talking about? What's your deal with China?

lxj034000
05-05-2011, 05:21 AM
My First Upper level exam & first exam with CAS, if previously joint exams do not count.
Guess the one made claims liability part must be a big fan of "being simple" or "being straight". For God's sake, how many times actuaries in practice compute ATA factors manually? Are we still living in 1980s when computers & software were still as remote as China's democracy?
Believe the exam should be more focusing on how to appropriately select ATA factors based on insurer's internal operation & external environment, instead of encouraging "super smart" individuals who can do easy math in their mind faster than calculators. If the latter was the case, why wouldn't they come to China and worship those kids who can be calculus experts at the age of 12.
Be careful!! This ia an actuarial forum. You are not supposed to offend others by your political or racial comments!

Inconceivable
05-05-2011, 08:06 AM
I took Exam 5B and managed to give (in my opinion) a decent answer for every question, but just barely; it was a mad race against the clock at the end, and I even blistered one of my fingers while writing. The problems were largely reasonable and recognizable, though I did not appreciate the requirement of simple averages. Anyone who has advanced this far up the exam hierarchy would presumably not have been able to do so had he/she not known how to calculate an arithmetic mean. I do not envy those who had to deal with that drudgery and the ratemaking material in one sitting.

Still, I think I am improving in my ability to complete these exams in the time provided. As with Exam 7 this sitting, I was able to answer all the questions, and I am more confident in my answers here than on Exam 7. An extra 15-30 minutes would have allowed me to provide excellent answers, but the ones I gave might do (I hope).

I agree entirely with those who question whether these exams appropriately test the skills needed to be a successful actuary. If anybody tried to churn out a real actuarial product at the rate the CAS expects us to solve exam problems, I would seriously question the quality. Are we supposed to be thoughtful, critically thinking professionals or frenzied button-pushing trained monkeys whose jobs can be outsourced to Microsoft Excel?

Sincerely,
G. Stolyarov II, ARe, AIS

Well said. But seems like it's something that will not change because that's how it's always been. Sigh. The exam proctor even chatted me up after the exam because I was writing so furiously and shaking my hand all the time because of the fatique. If I had 15-20 more minutes I could have guaranteed a pass - so what does that mean - I know the material well enough to do the work. But I might not pass the exam because I don't write that fast? I'm not saying give people unlimited time (some cap should be put on to prevent abuse) but I for the life of me don't get why the CAS makes them so you can't have an honest attempt at it.

Guest
05-05-2011, 09:43 AM
The length of this exam was ridiculous. You have to be a super number crunching ninja monkey to accurately do so many link ratios and do so in a reasonably presentable fashion.

shab
05-05-2011, 09:58 AM
I strongly agree that the time was not enough. I think they should put back few MCQ's to save some time from writing out your complete answer. I am reminding everyone to please fill out the survey, may be they would consider our feedback when determining the pass mark for this exam.

DrActuary
05-05-2011, 10:01 AM
where can we find the survey information? would it be possible to send in a link?

Vorian Atreides
05-05-2011, 10:03 AM
where can we find the survey information? would it be possible to send in a link?
Check this thread:

http://www.actuarialoutpost.com/actuarial_discussion_forum/showthread.php?t=216545

I have it stickied since this question keeps popping up in different threads.

If everyone can pass the word . . .

shab
05-05-2011, 10:09 AM
CAS Exam Survey.

http://www.casact.org/admissions/index.cfm?fa=surveys

apple
05-05-2011, 10:25 AM
I thought the questions were easy, but the test was really really long.

ditto... ah well, I know the info... but based on the ones I didn't get to due to time... I don't know that I've proved it...

p&cnewbie
05-05-2011, 10:36 AM
I agree that there were a lot of numbers involved. However, I use the data function on my TI-30XS Multiview to calculate ratios, which I think reduces time. I can calculate an entire triangle and then write it down all at once instead of going back and forth between my paper and calculator. The simple averages don't take me any longer than volume-weighted averages. It's about the same amount of numbers with parentheses in different places. Even with this practice, the test was too long for me.

what the...?
05-05-2011, 10:52 AM
welcome to upper levels...you are not fast if you haven't demonstrated speed by successfully completing atleast a few of the upper level exams under time pressure...i can sympathize with the feelings people have here, but that's what these are about.

I've taken other upper level exams and never felt as rushed as I did on this one. This exam was too long.

nonactuarialactuary
05-05-2011, 10:58 AM
I agree that there were a lot of numbers involved. However, I use the data function on my TI-30XS Multiview to calculate ratios, which I think reduces time. I can calculate an entire triangle and then write it down all at once instead of going back and forth between my paper and calculator. The simple averages don't take me any longer than volume-weighted averages. It's about the same amount of numbers with parentheses in different places. Even with this practice, the test was too long for me.

I've used my BA-II Plus for every exam to date, because I needed it for FM. For this exam, I finally went out and got a new calculator. I've heard a lot about the TI-30XS Multiview on these boards. I figured that I should look for that one because time was a big issue for me on both 5 and 6 last year. I went to all the usual stores, but couldn't find it anywhere. The closest I could find was a TI-34 Multiview, but I didn't see that one on the approved calculators list. Did you have to specially order this online?

I ended up settling with the TI-30XIIS, but I brought both calculators into the exam room. The TI-30XIIS seemed a good bit sturdier than the BA-II Plus. It was nice actually having faith that if I pressed the number 7, it would actually show up on the screen. Parentheses were a huge plus too. Because I bought it the day before the exam, I didn't have time to learn the features, so I actually flipped between the two calculators quite a bit, sometimes even during the same problem. None of this should matter, of course, but when the CAS continues to design ridiculously long exams, every second matters. For exam 6U in the fall, would you recommend that I get the TI-30XS Multiview and learn the features now, or are my calculators good enough?

Guest
05-05-2011, 11:15 AM
Sorry if already answered but is the pass mark for the full exam just pass mark for 5a + pass mark for 5b or determined separately? Clearly there's a huge advantage for taking only half an exam.

p&cnewbie
05-05-2011, 11:26 AM
Nobody knows (except perhaps the 'insiders') at this point.

enactuary
05-05-2011, 12:07 PM
Sorry if already answered but is the pass mark for the full exam just pass mark for 5a + pass mark for 5b or determined separately? Clearly there's a huge advantage for taking only half an exam.

True, but those who pass the whole exam have a huge advantage over those who pass the half exam in terms of credit gained. Even by passing 5b, you still get "caught" by everyone who passes the whole exam.

Also, the half exams may be easier, but the process isn't, especially for those who took 5b. Most of those folks had a crack at the old reserving exam. By not passing it, they have to take 1.5 exams to get the same credit, which may as well be counted as 2 exams (since taking 5b and 7 on consecutive days is foolish for most). This means it SHOULD be easier pass a half exam, since that's the ONLY advantage. Everything else is an advantage for the full-exam takers.

TRINIDON2K
05-05-2011, 12:29 PM
I should have said "f you" and just did VWA's instead off all year average... they'd probably take off a half point for that. Wasted sooooo much time with those annoying link ratios.

Thought it was an easy test but got killed on time. At least i'm not alone. My hand still hurts from writing so much so fast.

JOLE5646
05-05-2011, 12:40 PM
I don't think people realize how easy it is to do straight average on your calculator. Use the brackets.

straight avg: ((234/123)+(252/133)+(211/111))/3

weighted avg: (234+252+211)/(123+133+111)

It shouldn't take you any longer to do either method.

Number Theory
05-05-2011, 12:43 PM
You have to show the LD triangle and an example formula. That is what took so long.

p&cnewbie
05-05-2011, 12:45 PM
I always pick the calculation using two ratios, so it cuts down on writing but still shows process.

JOLE5646
05-05-2011, 12:50 PM
You have to show the LD triangle and an example formula. That is what took so long.

Since when do you have to show an example formula for the selection of link ratios?

I would just do:

12-24: x
24-36: y
36-48: z

etc... no formula provided because they already told you which method to use.

I remember those link ratios question have no more than 1 or 2 points assigned to them so don't provide too many details.

CASualty
05-05-2011, 01:11 PM
Since when do you have to show an example formula for the selection of link ratios?

I would just do:

12-24: x
24-36: y
36-48: z

etc... no formula provided because they already told you which method to use.

I remember those link ratios question have no more than 1 or 2 points assigned to them so don't provide too many details.

You are supposed to show your calculations . . . if you simply write down the values, it may (or may not) be enough for the CAS grader. Due to time constraints, when they asked for volume-weighted averages I would just give them my final ATA selections (since I think that is obvious enough). When they specifically asked for an average of the link ratios, I would show the entire ATA triangle and my selections (but wouldn't show any sample calculations since I think it's obvious enough).

Ceej
05-05-2011, 01:16 PM
Since when do you have to show an example formula for the selection of link ratios?


I actually had the same exact conversation with a co-worker that doesn't show the link ratio triangle and just lists the LDFs after doing the calculation in the calculator. I on the other hand, show all the intermidiate steps and was completely pressed for time.

I think reading that paragraph on the first page of the instructions scares the crap out of me and makes me show more work than I think is necessary. It notes (bolded and underlined for stylistic emphasis) that in order to receive full credit or to maximize partial credit that you should show calculations and formulas used to derive solutions. To me, that means showing the triangle calculation to arrive at the LDFs but I do see the point that it's such a small part of the process that to waste your time showing the work - I don't know what I gained (and more probable lost) over my coworker's method.

JOLE5646
05-05-2011, 01:22 PM
You are supposed to show your calculations . . . if you simply write down the values, it may (or may not) be enough for the CAS grader. Due to time constraints, when they asked for volume-weighted averages I would just give them my final ATA selections (since I think that is obvious enough). When they specifically asked for an average of the link ratios, I would show the entire ATA triangle and my selections (but wouldn't show any sample calculations since I think it's obvious enough).

That's a lot of time to spend on a 1 point question. You must be referring to the solutions provided by the study manuals, which showed example calculations for people to understand. I don't think the graders would require you to show example calculations. If they do, that's pretty silly. Even if they do, I would probably lose no more than 0.25 or 0.5 mark for not showing my work. I'd gladly sacrifice that for more time.

folasona
05-05-2011, 01:27 PM
I don't think people realize how easy it is to do straight average on your calculator. Use the brackets.

straight avg: ((234/123)+(252/133)+(211/111))/3

weighted avg: (234+252+211)/(123+133+111)

It shouldn't take you any longer to do either method.


notice how we're all talking about what buttons you should push and not push and what you may or may not have to show, but hardly any post-exam discussion is touching on the topics of how to actually understand the material or what you have may or may not have to know

I started out showing enough work to make sure that I would at least receive maximum partial credit on the problems, which led directly to making sure that I would lose out on whole credit for a whole bunch of other problems (most of which looked quite understandable and straightforward during the reading period...the first and last time I got a chance to glance at them)

From a grading standpoint, you'd think it would be easier to grade coherent, thought out, and legible writing, and organized triangles / rows / columns / diagrams / formulas and calculations than rather than illegible chicken scratch full of makeshift abreviations and accronyms, and random circles and squigly arrows trying to to make sense of numbers jumbled together like rush hour traffic or sprawled accross the page like food after a cafeteria foodfight (you know the type of thing that occasionally happens during the age students are first learning how to do ratios and simple averages...)

but then again, thankfully I'm not a grader so who knows...

Present Value
05-05-2011, 01:28 PM
In any upper level exam, follow this rule of thumb and you guys should be relatively close.

If your definition of showing "work" results in you leaving x points or more overall, then don't show the "work", else show "work".

Note I have said your definition of "work" and x points. Think of the exam as a constrained maximization problem with these two variables.

The ideal candidate will have a good "sick sense" about when he is showing too much "work" which is jeopardizing his ability to finish the exam. Too often exams are passed based on not how much you know but how little you leave. This may be sub-optimal, but this is the process and you have chosen to beat it.

So, practice harder on this aspect in the future. Also, I disagree with this not being needed in real life. For all those who say there is no time constraint in real life other than consulting...wait for a few years. In a few years time you will find yourself and your team managing scarce resources on competing projects. You have to determine resource allocation and you have to find the value of projects you choose over the ones you leave. IMO that is no different from what you are being asked to do on these exams

Adversely Selected
05-05-2011, 01:38 PM
I agree entirely with those who question whether these exams appropriately test the skills needed to be a successful actuary. If anybody tried to churn out a real actuarial product at the rate the CAS expects us to solve exam problems, I would seriously question the quality. Are we supposed to be thoughtful, critically thinking professionals or frenzied button-pushing trained monkeys whose jobs can be outsourced to Microsoft Excel?

Sincerely,
G. Stolyarov II, ARe, AIS

You said it better than I did.

I strongly agree that the time was not enough. I think they should put back few MCQ's to save some time from writing out your complete answer. I am reminding everyone to please fill out the survey, may be they would consider our feedback when determining the pass mark for this exam.

Yeah right, nothing has ever been done because of those surveys. Look at the historical results of those surveys. Everyone of them always says TOO LONG or not adequate coverage of the syllabus...yet, it seems to be getting worse. They use those surveys to laugh.

notice how we're all talking about what buttons you should push and not push and what you may or may not have to show, but hardly any post-exam discussion is touching on the topics of how to actually understand the material or what you have may or may not have to know



Very good point! Usually the discussion revolves around "How did you solve this?", not "Did anyone get confused on the 300 link ratio calcuations?"

FourKicks
05-05-2011, 01:42 PM
Very good point! Usually the discussion revolves around "How did you solve this?", not "Did anyone get confused on the 300 link ratio calcuations?"

wait til the exams come back in the mail.

Ceej
05-05-2011, 01:45 PM
I thought all the problems were pretty straight forward for Exam 5B. I didn't think that any of them were "oh gee, I don't know what to do here" or "hmm, what did they mean by this?" I thought last sitting of Exam 6 there were alot of grey questions and the corresponding discussion thread mirrored that by alot discussion about what was meant or how one interpretted certain questions. What I saw this sitting everything was pretty vanilla, just long hence the discussion around the button pushing. The only one that may have escaped that was the reinsurance one as I wasn't sure where in Friedland it speaks about this and answered it based on my old Exam 6 knowledge.

nonactuarialactuary
05-05-2011, 02:20 PM
Also, I disagree with this not being needed in real life. For all those who say there is no time constraint in real life other than consulting...wait for a few years. In a few years time you will find yourself and your team managing scarce resources on competing projects. You have to determine resource allocation and you have to find the value of projects you choose over the ones you leave. IMO that is no different from what you are being asked to do on these exams

Disagree 100%.

There's time constraints, and there's unnecessary time constraints. I'm in consulting, so I've seen a few 60-70 hour weeks in my time. Even in my busiest weeks, I face nothing remotely close to the CAS exam time limits. If I want a sandwich, I can still head down to the shop to get one. I'm not furiously tapping away at my computer until my hands go numb. The fact that the majority of the post-exam discussion is centering around which order to use your parenthesis in a calculator is evidence of a broken system. People aren't time constrained because they took too long to think about something that should have been internalized. If that were the case, you'd have a point. Instead, people are time constrained because a problem gives you seven years of data when the exact same question could be applied to four years of data. In any real world situation, calculating a triangle of link ratios for seven years of data takes only a few seconds longer than calculating a triangle of link ratios for four years of data. On an exam where you're punching everything in by hand one by one? Not so much.

The actual numerical values used in questions matter as well. If I see a triangle of losses valued at $20,000 @12 months and $25,000 @24 months, I can easily come up with the link ratio in my head and move on to the next part of the problem. Instead, we see triangles on the exam with losses valued at $20,674 @12 months and $25,746 @24 months. In the real world, the differences between the two scenarios are trivial. On the exams, the second one takes way more time. First, you need to actually punch the values into your crappy calculator to do the calculation. Second, you're left with a value that extends to multiple decimal places, so proper rounding techniques are required, none of which are adequately documented. I can demonstrate mastery of the subject with the first scenario just as well as I can with the second scenario - the only additional thing the second scenario tests is my button pressing speed. The scenarios are identical in the real world if you're doing calculations in Excel why should they matter on an exam?

I'm fine with time constraints, and I'm fine with them making the exam longer, but when the majority of the length comes from calculations that are long for the sake of being long without adding any additional syllabus coverage, I question the usefulness of the exam system altogether.

Guest
05-05-2011, 02:22 PM
In any upper level exam, follow this rule of thumb and you guys should be relatively close.

If your definition of showing "work" results in you leaving x points or more overall, then don't show the "work", else show "work".

Note I have said your definition of "work" and x points. Think of the exam as a constrained maximization problem with these two variables.

The ideal candidate will have a good "sick sense" about when he is showing too much "work" which is jeopardizing his ability to finish the exam. Too often exams are passed based on not how much you know but how little you leave. This may be sub-optimal, but this is the process and you have chosen to beat it.

So, practice harder on this aspect in the future. Also, I disagree with this not being needed in real life. For all those who say there is no time constraint in real life other than consulting...wait for a few years. In a few years time you will find yourself and your team managing scarce resources on competing projects. You have to determine resource allocation and you have to find the value of projects you choose over the ones you leave. IMO that is no different from what you are being asked to do on these exams

IMHO, the process is not consistent. In one year they stress harder problems with more gray areas, then the next it's number punching easy problems so it is somewhat random in what type of candidate succeeds. A perfect candidate that has mastered both types of tests will of course succeed but if you happen to get the "right" test even if you aren't the perfect candidate you will do just as well.

All in all, every student chose this path knowing the risks and imperfections built into the process, and that is definitely true. No one is forcing you to deal with all this and it's curved so everyone suffers. I'm also aware that graders and committee members are working 100% pro bono. You definitely have to be grateful for that and a system supported by volunteers will naturally have limitations that need to be accepted to some degree. That being said, not all the complaints are 100% invalid. The fact that you can't get your answers back or the rubrick doesn't improve the credibility of the exams.

Another thing frustrating personally is the typical bell curve that you see with other types of exams you don't see in CAS exams. There is a relatively tight band between the pass mark and the 95th percentile. Take exam 5 2009. The pass mark which is at the 54th percentile, was 55. The 95th percentile was a mere 10 points higher at 65. That's the difference between a superb top tier candidate and middle of the road 54th percentile candidate. How much of that is luck and volatility due to random fluctuation of scores (especially under extraordinary time pressure)? If someone gets the "right" test I can easily see them nailing an additional 5, to maybe even 10, points.

/rant

jjb3113
05-05-2011, 02:58 PM
I've used my BA-II Plus for every exam to date, because I needed it for FM. For this exam, I finally went out and got a new calculator. I've heard a lot about the TI-30XS Multiview on these boards. I figured that I should look for that one because time was a big issue for me on both 5 and 6 last year. I went to all the usual stores, but couldn't find it anywhere. The closest I could find was a TI-34 Multiview, but I didn't see that one on the approved calculators list. Did you have to specially order this online?

I was paranoid that my TI-30XIIS was going to quit on me mid-exam since I have had it for 5-6 years, I went to Best Buy to buy a new one. I saw the TI-30XS Multiview there as well, but opted for the caclulator that I have become accustomed to.

SOAPhobic
05-05-2011, 04:43 PM
There's time constraints, and there's unnecessary time constraints. I'm in consulting, so I've seen a few 60-70 hour weeks in my time. Even in my busiest weeks, I face nothing remotely close to the CAS exam time limits. If I want a sandwich, I can still head down to the shop to get one. I'm not furiously tapping away at my computer until my hands go numb. The fact that the majority of the post-exam discussion is centering around which order to use your parenthesis in a calculator is evidence of a broken system. People aren't time constrained because they took too long to think about something that should have been internalized. If that were the case, you'd have a point. Instead, people are time constrained because a problem gives you seven years of data when the exact same question could be applied to four years of data. In any real world situation, calculating a triangle of link ratios for seven years of data takes only a few seconds longer than calculating a triangle of link ratios for four years of data. On an exam where you're punching everything in by hand one by one? Not so much.


:yawn: There's wordiness and there's unnecessary wordiness. You fall in the latter. Now I know the real reason you whine.


The actual numerical values used in questions matter as well. If I see a triangle of losses valued at $20,000 @12 months and $25,000 @24 months, I can easily come up with the link ratio in my head and move on to the next part of the problem. Instead, we see triangles on the exam with losses valued at $20,674 @12 months and $25,746 @24 months. In the real world, the differences between the two scenarios are trivial. On the exams, the second one takes way more time.

If you fail because of this second that's "taken away" from you, you fail not because nasty numbers wasted your time, you simply didn't study for a 10.


First, you need to actually punch the values into your crappy calculator to do the calculation. Second, you're left with a value that extends to multiple decimal places, so proper rounding techniques are required, none of which are adequately documented. I can demonstrate mastery of the subject with the first scenario just as well as I can with the second scenario - the only additional thing the second scenario tests is my button pressing speed. The scenarios are identical in the real world if you're doing calculations in Excel why should they matter on an exam?

I'm fine with time constraints, and I'm fine with them making the exam longer, but when the majority of the length comes from calculations that are long for the sake of being long without adding any additional syllabus coverage, I question the usefulness of the exam system altogether.

I've seen people pass an actuarial exam by skipping 1/4 of the points on the exam. I sense somebody's looking for scrap points to pass.

2M
05-05-2011, 04:48 PM
:iatp:

Present Value
05-05-2011, 04:56 PM
If a candidate is prepared you can attempt 90% or less of the paper in the allotted time, take out all the "speed test" issues, number crunching issues, and still pass.

2M
05-05-2011, 05:00 PM
Part of the purpose of these exams is to test speed as well as accuracy, not just the accuracy. There are ways to get through all the problems on this exam in the time allotted. Something I am learning the hard way as well btw :wall:


Prerequisite: Know the material really well. Studying with your sights set on 100% not 70%. Yes, it is possible to skip entire sections of the exams and still pass, I have seen people do this although I wouldn't recommend it.

Assuming you know the material, for this test, then you need to become adept at writing down only what you need to and calculating only what you need to.

A few hypothetical examples:

If you are given triangles of paid claims and reported claims and you need case reserves for a berquist-sherman problem. No need to recopy the the paid and reported triangles. Just calculate the case.

If you need to compute indiviual ATA's as many of us did yesterday then do so but write down only the ATA's not the undelying triangle data that goes with it.

If you are asked to calculate ultimate values for AY 2009, no need to calculate AY 2008 and 2007 too.

Efficiency could potentially save a couple minutes off each problem which adds up over the course of the exam.

enactuary
05-05-2011, 05:33 PM
Efficiency could potentially save a couple minutes off each problem which adds up over the course of the exam.

I agree with everything you said, but the frantic nature of the exam (for me at least) increases the likelihood of missing an important detail and actually costing more time than if the question were approached at a comfortable pace.

I always have speed issues, and although I didn't finish the exam, I got a lot closer than usual. I believe part of that was due to taking an extra couple of seconds to pay more attention to what the question was asking. I have shot myself in the foot before completing triangles that weren't necessary had I read the question more carefully.

Use the reading period to look for potential time traps like that. I've lost a lot more time from those situations than from doing the mechanical work too slowly. Efficiency is about more than just pure speed.

The lesson: Don't be in such a hurry to get on the road that you forget to plan your route. You'll pay the price when you hit a dead end and have to double back.

nonactuarialactuary
05-05-2011, 05:47 PM
:yawn: There's wordiness and there's unnecessary wordiness. You fall in the latter. Now I know the real reason you whine.


If you fail because of this second that's "taken away" from you, you fail not because nasty numbers wasted your time, you simply didn't study for a 10.



I've seen people pass an actuarial exam by skipping 1/4 of the points on the exam. I sense somebody's looking for scrap points to pass.

Meh. For the record, I actually felt pretty good on the timing. I only left one 2.5 point question and a few of the sub-points entirely blank, so I feel that I have a good chance to pass this thing. It's just frustrating to see triangles with seven years of data, but no Berquist-Sherman adjustment problems for claim settlement patterns. It's frustrating to calculate the same exact reported loss ratio 15 times when a disposal rate frequency/severity problem doesn't get asked. Entire swaths of the syllabus were skipped in favor of unnecessarily long division problems. I think you'd have a better shot distinguishing qualified versus unqualified candidates by including more of the syllabus. Instead, the whole exam amounted to nothing more than a race to type numbers into a calculator.

Number Theory
05-05-2011, 06:32 PM
There were no BS adjustment problems? I thought that there was one where you had to linearly interpolate?

Inconceivable
05-05-2011, 07:11 PM
There were no BS adjustment problems? I thought that there was one where you had to linearly interpolate?

OK, now I'm getting confused. I thought there was a BS adjustment one but it was for case o/s? At this point my memory is failing me pretty badly though.

Number Theory
05-05-2011, 07:16 PM
You're probably right. It said to test the data to see if adjustments need to be made (which Ididnt have time for), but in the instructions it said assume uniform payment patterns so I thought that should mean linear interpolation... maybe my memory is gone as well... cause I think I remember doing both adjustments at some point during the exam.

kyleucf
05-05-2011, 07:17 PM
there was a case o/s adj, and alot of questions about reserve adequecy and speed of settlement...

PhildeTruth
05-05-2011, 07:18 PM
:yawn: There's wordiness and there's unnecessary wordiness. You fall in the latter. Now I know the real reason you whine.


If you fail because of this second that's "taken away" from you, you fail not because nasty numbers wasted your time, you simply didn't study for a 10.



I've seen people pass an actuarial exam by skipping 1/4 of the points on the exam. I sense somebody's looking for scrap points to pass.

:iatp:

kyleucf
05-05-2011, 07:27 PM
what did you all put for that glm output.

Number Theory
05-05-2011, 07:34 PM
probably wrote more than I should have for that one. Not predictive; that was almost straight out of the last appendix

ajoh321
05-05-2011, 07:50 PM
I agree on the GLM - just noise - not a predictor. I thought the exam was long - certainly knew enough to score over 70 but not sure I did. Not at all clear to me why they think it is meaningful to test ability to add and divide lots of numbers. It means nothing at all IMO and I think there was a lot of the syllabus that was untested. I think the quality of these exams is low. I am hopeful but if I didn't get it, I'll be back next year!

Didn't do enough speed practice and practice exams. Went through all the SOA and CAS math tests and averaged a 9 - now I have a hard time getting a 6!!

earthsunmoonsky
05-05-2011, 07:50 PM
OK, now I'm getting confused. I thought there was a BS adjustment one but it was for case o/s? At this point my memory is failing me pretty badly though.

The B-S problem was using case reserve adequacy. It gave a trend of 10% and a few triangles.

What did you all put for the 5% trend making higher or lower estimates? I put lower but some lame reason about how the avg case outstanding would be lower....

Bobby
05-05-2011, 07:53 PM
I put lower. Now that you have your exam booklet back, would it actually be lower? I didn't have time to actually calculate it. My reasoning was that the LDF's would be smaller.

ajoh321
05-05-2011, 07:55 PM
I also wrote a short novel for the question about overlap fallacy. It is so stupid but the paper itself doesn't give a very good explanation and the diagram they use seems to imply that development and trend apply to different time periods. I think that misses the point that trend has the nature of inflation over time and development has to do with undoing an estimate. Easy to see that there is no overlap if you consider the cases where either trend or development is not indicated. Probably wasted 5 minutes on that problem and probably wont get full credit.

earthsunmoonsky
05-05-2011, 07:56 PM
I put lower. Now that you have your exam booklet back, would it actually be lower? I didn't have time to actually calculate it. My reasoning was that the LDF's would be smaller.

I calculated it on the exam to make sure. It is lower. I just wanted to know if anyone had a less lame reason than what I put down.

Number Theory
05-05-2011, 08:02 PM
Can't remember what I put. Probably just put "lower" or "higher" due to time.

snoo
05-05-2011, 08:07 PM
I calculated it on the exam to make sure. It is lower. I just wanted to know if anyone had a less lame reason than what I put down.

I put the same reason as you...

earthsunmoonsky
05-05-2011, 08:07 PM
I put the same reason as you...

Nice. I wrote about 3 sentences, or maybe just one really long run-on sentence that talked in circles.

Vorian Atreides
05-05-2011, 08:12 PM
The B-S problem was using case reserve adequacy. It gave a trend of 10% and a few triangles.

What did you all put for the 5% trend making higher or lower estimates? I put lower but some lame reason about how the avg case outstanding would be lower....

I put lower. Now that you have your exam booklet back, would it actually be lower? I didn't have time to actually calculate it. My reasoning was that the LDF's would be smaller.
This. Smaller trend ==> larger case o/s ==> larger adjusted reported losses ==> smaller LDFs ==> smaller IBNR estimates.

earthsunmoonsky
05-05-2011, 08:18 PM
This. Smaller trend ==> larger case o/s ==> larger adjusted reported losses ==> smaller LDFs ==> smaller IBNR estimates.

Good answer! Just realized I said I put smaller case o/s. Hopefully if I actually put that on the exam they forgive my blatant stupidity as my example showed otherwise.

lxj034000
05-05-2011, 08:19 PM
probably wrote more than I should have for that one. Not predictive; that was almost straight out of the last appendix

I said something like only "Sprinkler" is significant and should receive a credit, but b/c the exposure of Sprinkler is small, the variable is not very efficient.

Bobby
05-05-2011, 08:20 PM
This. Smaller trend ==> larger case o/s ==> larger adjusted reported losses ==> smaller LDFs ==> smaller IBNR estimates.

I actually have you to thank for that point. You replied to someone in this very forum regarding how larger policy limits mean smaller LDF's. Also a big thanks to the person who posted that question. Wouldn't have guessed that without reading the AO. Even if I don't pass, I owe you a beer. :toast:

Vorian Atreides
05-05-2011, 08:26 PM
I actually have you to thank for that point. You replied to someone in this very forum regarding how larger policy limits mean smaller LDF's. Also a big thanks to the person who posted that question. Wouldn't have guessed that without reading the AO. Even if I don't pass, I owe you a beer. :toast:
I'll likely take you up on that. :-) Perhaps at an Exam 6 in-person seminar . . .

Bobby
05-05-2011, 08:33 PM
I'll likely take you up on that. :-) Perhaps at an Exam 6 in-person seminar . . .

Yeah I'll probably be at the TIA seminar myself for 6 this fall (didn't do it for 5 because I thought this exam was going to be a piece of cake... hah!). So I'll see you then.

Number Theory
05-05-2011, 08:38 PM
hmmm... I remember what I put for that one now. I put that they would be the same... thought it was a trick question.

Number Theory
05-05-2011, 08:43 PM
I still dont see how that makes the LDFs change. I would have to calculate that out on an exam to get th points.

Bobby
05-05-2011, 08:50 PM
I still dont see how that makes the LDFs change. I would have to calculate that out on an exam to get th points.

http://www.actuarialoutpost.com/actuarial_discussion_forum/showthread.php?t=216211

Number Theory
05-05-2011, 08:55 PM
But you are applying the change to the whole triangle not just the latest diagonal.

snoo
05-05-2011, 08:58 PM
This. Smaller trend ==> larger case o/s ==> larger adjusted reported losses ==> smaller LDFs ==> smaller IBNR estimates.

Facepalm!

Vorian Atreides
05-05-2011, 09:17 PM
But you are applying the change to the whole triangle not just the latest diagonal.
True, but you're using the new triangle to create the LDFs. It's these adjusted LDFs that then get applied to the latest diagonal.

B-S method isn't a method to estimate ultimate losses, just an adjustment technique to account for changes in operations.

Number Theory
05-05-2011, 09:21 PM
Oh so the LDFs are lower by a factor of 1.05, but so is your ultimate for all AYs, so the IBNR is the same. That's what I would have said probably because that's how I am picturing the results without actually calculating to see.

Vorian Atreides
05-05-2011, 09:25 PM
Oh so the LDFs are lower by a factor of 1.05, but so is your ultimate for all AYs, so the IBNR is the same. That's what I would have said probably because that's how I am picturing the results without actually calculating to see.
Reported on the latest diagonal isn't any different from the original data--only the internal cells of the diagonal are different.

So you're applying smaller LDFs to the same numbers on the latest diagonal which will result in a smaller estimate of ultimate claims as before . . . but you're subtracting the same reported losses in both cases . . . ergo, smaller IBNR estimate.

Number Theory
05-05-2011, 09:29 PM
True so it takes me about 5 minutes to reason that out in my head. The only way anyone could see something like that instantaneously is if they thought through the answer to that question in their heads before hand. So I spent about .5 of a second writing my answer and moved on.

earthsunmoonsky
05-05-2011, 09:31 PM
For the frequency severity problem, did you only use the severity for 2010 since they only asked for ultimate claims for 2010? To refresh your memory they gave total closed counts and counts with no pay. I made an adjusted triangle x counts of no pay calculated age to age. Then used age to ult to get ult 2010 counts and multiplied by 2010 severity which was given. Sound familiar?

Number Theory
05-05-2011, 09:32 PM
I graded myself based on an all or nothing scale and gave myself a 0/3.5 for the BS question. Can't remember why but I feel like I did linear interpolation somewhere in this exam and it must have been this question!

Number Theory
05-05-2011, 09:34 PM
for that one i calculated an new closed with pay triangle
then I took the tail and trended each row to 2010 levels
then summed the severity divided by the claim counts with pay.

earthsunmoonsky
05-05-2011, 09:37 PM
for that one i calculated an new closed with pay triangle
then I took the tail and trended each row to 2010 levels
then summed the severity divided by the claim counts with pay.

This could be me being stupid, but they didn't give us a trend. Just the 2 triangles, and severity for each year (2007-2010). Can you explain how/why you did this?

Vorian Atreides
05-05-2011, 09:38 PM
For the frequency severity problem, did you only use the severity for 2010 since they only asked for ultimate claims for 2010? To refresh your memory they gave total closed counts and counts with no pay. I made an adjusted triangle x counts of no pay calculated age to age. Then used age to ult to get ult 2010 counts and multiplied by 2010 severity which was given. Sound familiar?
Yep . . . only I was running out of time and knew that I couldn't do all of that in the time remaining.

So I just assumed an ultimate claim count for 2010 and finished the problem.

I'll see sometime in July how well that tactic fared.

earthsunmoonsky
05-05-2011, 09:41 PM
Yep . . . only I was running out of time and knew that I couldn't do all of that in the time remaining.

So I just assumed an ultimate claim count for 2010 and finished the problem.

I'll see sometime in July how well that tactic fared.

Actually not a bad idea if you are short on time. Good luck! You def deserve a pass!

snoo
05-05-2011, 09:47 PM
For the frequency severity problem, did you only use the severity for 2010 since they only asked for ultimate claims for 2010? To refresh your memory they gave total closed counts and counts with no pay. I made an adjusted triangle x counts of no pay calculated age to age. Then used age to ult to get ult 2010 counts and multiplied by 2010 severity which was given. Sound familiar?

I subtracted the 2 triangles to get closed with pay. Then calculated cdfs to get ultimate counts. Hesitated for a moment where i thought about taking the avg of the projected severities but i didnt!

Number Theory
05-05-2011, 09:48 PM
Maybe I am thinking of another problem. I have a very shoddy memory on these things because pretty much I don't read any of the questions after about a quarter in because I feel as if I am running out of time. I think it was either 26 or 28 (on full exam)

Number Theory
05-05-2011, 09:49 PM
I think that you had to calculate the Closed with pay triangles because the severity was closed with pay severity

earthsunmoonsky
05-05-2011, 10:02 PM
Thanks guys! Last one, then its off to have some margaritas:

last question on the exam about the IBNR:

a) Dev Rptd and Pd had correct IBNR. Why?
b) Dev pd, Expected clms and Paid BF had correct IBNR. Why?
c) Rptd dev, Expected clms and Rptd BF had correct IBNR. Why?
d) Last one had all different IBNRs. Briefly discuss a next step the actuary should take to understand the difference in results?

I put:
a) LR increase
b) Case O/S increase
c) ??
d) look at ECRs, year to year changes, industry knowledge in context with management discussion, etc, etc to see what makes sense to do.

Kongo
05-05-2011, 10:05 PM
C is an increase in settling cases, or payment patterns.

And for d you are right, I didn't have time to write multiple things so just wrote, "before a retro test to see which performed better" I hope that gets me credit.

Kongo
05-05-2011, 10:06 PM
"perfrom retroactive test"

CASualty
05-06-2011, 12:02 AM
This. Smaller trend ==> larger case o/s ==> larger adjusted reported losses ==> smaller LDFs ==> smaller IBNR estimates.

I think that is exactly what I put, implied arrows and all.

JOLE5646
05-06-2011, 12:04 AM
I subtracted the 2 triangles to get closed with pay. Then calculated cdfs to get ultimate counts. Hesitated for a moment where i thought about taking the avg of the projected severities but i didnt!

I think you can take the average of all year's severity only if they are at current year's level (because of trending). Since they didn't say it was, I just used AY 2010 ultimate severity.

Adversely Selected
05-06-2011, 12:14 AM
I was paranoid that my TI-30XIIS was going to quit on me mid-exam since I have had it for 5-6 years, I went to Best Buy to buy a new one. I saw the TI-30XS Multiview there as well, but opted for the caclulator that I have become accustomed to.

Everyone at my sitter had those calculators. I've never seen them before. I had my trusty TI-30XIIS. Before the exam I thought I missed some notification of which calculator to use. I ordered one after the exam in order to test it out.

I also wrote a short novel for the question about overlap fallacy. It is so stupid but the paper itself doesn't give a very good explanation and the diagram they use seems to imply that development and trend apply to different time periods. I think that misses the point that trend has the nature of inflation over time and development has to do with undoing an estimate. Easy to see that there is no overlap if you consider the cases where either trend or development is not indicated. Probably wasted 5 minutes on that problem and probably wont get full credit.

I think I put a simple answer straight from the text about severity being thought to be double counted in the calculation whereas in reality trend is bringing losses to the future period and LDFs are bringing those losses to their ultimate value.

CASualty
05-06-2011, 12:20 AM
True so it takes me about 5 minutes to reason that out in my head. The only way anyone could see something like that instantaneously is if they thought through the answer to that question in their heads before hand. So I spent about .5 of a second writing my answer and moved on.

Actually, this is the fundamental reason why the B/S Case O/S Adjustment is applied in the first place. If you actually understand what the B/S Method is accomplishing (and not just memorizing the calculation), then this question is actually pretty easy.

Think of it this way -- in Friedland, they make it very clear that when Case O/S Adequacy is increasing, then the Reported Development Method will overstate IBNR. If you apply the B/S method to remove the effect of the Case O/S strengthening, then you are no longer overstating the IBNR (i.e. the IBNR estimate will decrease).

Number Theory
05-06-2011, 08:30 AM
Actually, this is the fundamental reason why the B/S Case O/S Adjustment is applied in the first place. If you actually understand what the B/S Method is accomplishing (and not just memorizing the calculation), then this question is actually pretty easy.

Think of it this way -- in Friedland, they make it very clear that when Case O/S Adequacy is increasing, then the Reported Development Method will overstate IBNR. If you apply the B/S method to remove the effect of the Case O/S strengthening, then you are no longer overstating the IBNR (i.e. the IBNR estimate will decrease).

Yes I totally get that, but I have to think about it like this during the exam...
(1) OK what happens to case when I apply the 10% trend- ok average case is reduced.
(2) what happens when I apply a 5% trend- ok average case is reduced but not by as much as before.
(3) So that means total case reserves are higher.
(4) So now I add them to the same paids.
(5) And end up with smaller reporteds
(6) Which means my LDFs are smaller.
(7) which means my IBNR is smaller.

Too much thinking for .5 to 1 point IMO

Bobby
05-06-2011, 08:34 AM
Yes I totally get that, but I have to think about it like this during the exam...
(1) OK what happens to case when I apply the 10% trend- ok average case is reduced.
(2) what happens when I apply a 5% trend- ok average case is reduced but not by as much as before.
(3) So that means total case reserves are higher.
(4) So now I add them to the same paids.
(5) And end up with smaller reporteds
(6) Which means my LDFs are smaller.
(7) which means my IBNR is smaller.

Too much thinking for .5 to 1 point IMO

Avg case is increased.

1 / 1.1 < 1 / 1.05

Inconceivable
05-06-2011, 08:38 AM
Man I effed that question up too. I am going to fail. I said that the average case will not be reduced as much as they will under the 10% trend, leading to paid + case o/s larger under the 5% scenario than under the 10% leading to higher reporteds and higher CDF ==> higher IBNR. WTF!?!?!??!?

JasonScandopolous
05-06-2011, 08:55 AM
I also wrote a short novel for the question about overlap fallacy. It is so stupid but the paper itself doesn't give a very good explanation and the diagram they use seems to imply that development and trend apply to different time periods. I think that misses the point that trend has the nature of inflation over time and development has to do with undoing an estimate. Easy to see that there is no overlap if you consider the cases where either trend or development is not indicated. Probably wasted 5 minutes on that problem and probably wont get full credit.

I mean, the first time I read the "overlap fallacy" section in the textbook, I thought they were going to talk about how there *IS* an overlap between trend and development. The thing is, there's an overlap if you're talking about calendar year trends, and there is not an overlap if you're talking about accident year trends. Loss dev is obviously seperate things if you realize that they mean AY (or PY) trends in the data. On the exam, I said something like "They are not the same thing. Loss trend re-states claims as if they had happened in the future, at future cost levels. Loss development brings claims to their ultimate value, regardless of what cost level they are at. No overlap exists".

Barnett and Zehnworth (2004?) discusses why calendar year trends *DO* overlap with loss development and cause problems. I thought that was a good paper (old exam 6 syllabus) because most trends are, in reality, CY trends (e.g. inflation). I found it very odd for them to teach us this fact on the old exam 6, and then turn around and dumb it down for the new exam 5.

Inconceivable
05-06-2011, 09:09 AM
Actually, this is the fundamental reason why the B/S Case O/S Adjustment is applied in the first place. If you actually understand what the B/S Method is accomplishing (and not just memorizing the calculation), then this question is actually pretty easy.

Think of it this way -- in Friedland, they make it very clear that when Case O/S Adequacy is increasing, then the Reported Development Method will overstate IBNR. If you apply the B/S method to remove the effect of the Case O/S strengthening, then you are no longer overstating the IBNR (i.e. the IBNR estimate will decrease).

I read that as what will the IBRN be under the 5% assumption compared to under the 10% assumption. I said that it will be higher. I am appealing this one if I get it wrong because I think it can be read multiple ways.

snoo
05-06-2011, 09:09 AM
For the frequency severity problem, did you only use the severity for 2010 since they only asked for ultimate claims for 2010? To refresh your memory they gave total closed counts and counts with no pay. I made an adjusted triangle x counts of no pay calculated age to age. Then used age to ult to get ult 2010 counts and multiplied by 2010 severity which was given. Sound familiar?

Man I effed that question up too. I am going to fail. I said that the average case will not be reduced as much as they will under the 10% trend, leading to paid + case o/s larger under the 5% scenario than under the 10% leading to higher reporteds and higher CDF ==> higher IBNR. WTF!?!?!??!?

I even calculated it and botched it! This thread gives me the :ohnoes:

Inconceivable
05-06-2011, 09:51 AM
I even calculated it and botched it! This thread gives me the :ohnoes:

I didn't have time for it. :(

nonactuarialactuary
05-06-2011, 09:56 AM
I think that you had to calculate the Closed with pay triangles because the severity was closed with pay severity

Yep - that sounds vaguely familiar. I think I took the subtracted the two triangles they gave (total closed counts and closed without payment) to get closed with pay counts, projected them to ultimate for AY2010, and multiplied by the AY2010 severity. There may have been an exposure trend or something with this problem too, but I can't remember.

Staky41
05-06-2011, 10:00 AM
This. Smaller trend ==> larger case o/s ==> larger adjusted reported losses ==> smaller LDFs ==> smaller IBNR estimates.

ugh I just realized that in my hurry on the exam I was predicting higher ultimate losses.... I had this exact same reasoning but ==> higher ultimate reported losses :/

Staky41
05-06-2011, 10:03 AM
Reported on the latest diagonal isn't any different from the original data--only the internal cells of the diagonal are different.

So you're applying smaller LDFs to the same numbers on the latest diagonal which will result in a smaller estimate of ultimate claims as before . . . but you're subtracting the same reported losses in both cases . . . ergo, smaller IBNR estimate.

nevermind... as I'm reading through this, I am realizing that my last second scratches weren't correct at all... this makes sense. Good job.

earthsunmoonsky
05-06-2011, 11:42 AM
Thanks guys! Last one, then its off to have some margaritas:

last question on the exam about the IBNR:

a) Dev Rptd and Pd had correct IBNR. Why?
b) Dev pd, Expected clms and Paid BF had correct IBNR. Why?
c) Rptd dev, Expected clms and Rptd BF had correct IBNR. Why?
d) Last one had all different IBNRs. Briefly discuss a next step the actuary should take to understand the difference in results?

I put:
a) LR increase
b) Case O/S increase
c) ??
d) look at ECRs, year to year changes, industry knowledge in context with management discussion, etc, etc to see what makes sense to do.

Can anyone else confirm these answers? Thanks!

Inconceivable
05-06-2011, 11:53 AM
Can anyone else confirm these answers? Thanks!

That's what I put. I did have an answer for the ??? but can't remember it.

earthsunmoonsky
05-06-2011, 12:00 PM
That's what I put. I did have an answer for the ??? but can't remember it.


That makes two of us. I think the answer is spped up in payment patterns but I have no idea what I wrote on the exam. By that point I think my handwriting looked like a drunk doctor's signature.

Vorian Atreides
05-06-2011, 12:02 PM
Thanks guys! Last one, then its off to have some margaritas:

last question on the exam about the IBNR:

a) Dev Rptd and Pd had correct IBNR. Why?
b) Dev pd, Expected clms and Paid BF had correct IBNR. Why?
c) Rptd dev, Expected clms and Rptd BF had correct IBNR. Why?
d) Last one had all different IBNRs. Briefly discuss a next step the actuary should take to understand the difference in results?

I put:
a) LR increase
b) Case O/S increase
c) ??
d) look at ECRs, year to year changes, industry knowledge in context with management discussion, etc, etc to see what makes sense to do.

C is an increase in settling cases, or payment patterns.

And for d you are right, I didn't have time to write multiple things so just wrote, "perform a retro test to see which performed better" I hope that gets me credit.
Agree with c being a change (more specifically, a speed up) in settlement of claims.

As for d, I don't think perform retro testing is exactly what was looked for--more of which diagnostics should be done to see why there's the discrepancy with Pd Dev and Pd B-F. However, since it's a 1/4 point, I'll bet they'll give you credit (I would anyway, and if they don't I doubt that it's worth trying to appeal unless you have other problems you're going to appeal as well).

I think that esms' answer isn't that bad, either and may generate positive points (with same caveat mentioned for Kongo's response).

My answer to part d (and this isn't the only correct answer, I'm sure):
Check with Claims Dept to see if an event happened that slowed down claims processing.

Reasoning (not written on the Exam):
Since the paid estimates are lower from the other estimates, there's a chance that there's been a slow down in claim processing . . . checking with Claims to see if that might be the case is the quickest way to determine this.

Oddly enough, this is pretty much straight from Friedland (I'll need to look up the page reference though . . . I'll see if I can find it and edit this post accordingly).

Vorian Atreides
05-06-2011, 12:03 PM
That makes two of us. I think the answer is spped up in payment patterns but I have no idea what I wrote on the exam. By that point I think my handwriting looked like a drunk doctor's signature.
I've seen a drunk doctor's signature.

I've seen the same doctor's sober signature.

There wasn't much difference.






;-)

Inconceivable
05-06-2011, 12:12 PM
I think I did put speedup in payment patters so I should be ok on this one. For once.

Inconceivable
05-06-2011, 01:36 PM
ugh I just realized that in my hurry on the exam I was predicting higher ultimate losses.... I had this exact same reasoning but ==> higher ultimate reported losses :/

Anyone remember the point value of this part of the question? I'm guessing .5?

Vorian Atreides
05-06-2011, 01:50 PM
Anyone remember the point value of this part of the question? I'm guessing .5?
0.75 pts

earthsunmoonsky
05-06-2011, 01:59 PM
Thanks guys! I'm coming out at about 80% with a conservative grading estimate so hopefully this will be one of the last times I'm on this board. See you over in the 6 forum in July!

JOLE5646
05-06-2011, 02:33 PM
Thanks guys! I'm coming out at about 80% with a conservative grading estimate so hopefully this will be one of the last times I'm on this board. See you over in the 6 forum in July!

You do that too? I grade myself on every question based on the confidence of my answers. I practiced this with the practice exams and my estimates are very close to the actual. I'm sitting pretty comfortably in the mid 80s.

Inconceivable
05-06-2011, 03:00 PM
0.75 pts

Damn.

CASualty
05-06-2011, 03:52 PM
I read that as what will the IBRN be under the 5% assumption compared to under the 10% assumption. I said that it will be higher. I am appealing this one if I get it wrong because I think it can be read multiple ways.

The wording was:

"Discuss whether changing the annual severity trend given above from 10% to 5% would produce a higher or lower ultimate claims estimate under the Berquist-Sherman technique for case reserve adequacy".

I think it seems pretty clear?

CASualty
05-06-2011, 03:57 PM
Yep - that sounds vaguely familiar. I think I took the subtracted the two triangles they gave (total closed counts and closed without payment) to get closed with pay counts, projected them to ultimate for AY2010, and multiplied by the AY2010 severity. There may have been an exposure trend or something with this problem too, but I can't remember.

That's what I did as well. There was nothing about exposure trend.

The question gave you:
- Triangle: Cumulative Total Closed Claim Counts
- Triangle: Cumulative Claim Counts Closed with No Payment
- Projected Ultimate Severity per Claim Closed with Payment
- Assume no development after 48MO
- Use all-year volume-weighted

The question asked:
Use the Freq/Sev techinique to estimate the ultimate claim amount for AY2010.

Romas
05-09-2011, 10:53 AM
My impression that the exam was a bit too long.
During the prior reading, I identified where the Reserving part begins (from question 21 or 22), folded the angle of the page and told myself, that when 2 hours have passed, skip any remaining ratemaking questions and go to reserving. So, I had to skip some questions here and never had time to return to them. I also didn't have to time to address a couple of reserving questions, either.

untungl
05-09-2011, 05:05 PM
Last year I found myself scrambling to finish 6. I probably left 3 problems blank, or mostly blank.

I made some changes to my test taking strategy this time around though, and I think they helped. I finished with a few minutes to spare after looking back at some of the problems I earmarked.

Changes:
No restroom break (Thought I was gonna explode)
No snacks (Thought I was gonna die)
Only enough water to stay alive (Ok)

Ceej
05-09-2011, 07:50 PM
After taking the weekend to have fun in the sun, I took the survey.

Just a reminder - the survey does close two weeks after the exam. So, if you haven't done so try to do it before you forget completely.

JasonScandopolous
05-09-2011, 09:16 PM
Changes:
No restroom break (Thought I was gonna explode)
No snacks (Thought I was gonna die)
Only enough water to stay alive (Ok)

I dont know about anybody else, but I am frantically studying the day before, frantically cramming as much stuff into my mind as possible right before the test (simple memorization type stuff), and work as fast as possible the entire time. I have no idea why anybody would plan going to the bathroom (say this is 5 minutes: you just sacrified 1.5-2 points on the exam potentially), or need to; would you rather have a full bladder for an hour or would you rather study 300 hours and lose thousands of dollars of salary? Kinda beating a dead horse to say this to you since you already agree at this point, but I don't know why anybody does it...

Number Theory
05-09-2011, 09:23 PM
I am actually quite surprised that I didn't have to leave during the exam to use the restroom. The other two taking the exam with me each left twice (I think). They didn't finish the exam and neither did I.

FourKicks
05-09-2011, 11:27 PM
I dont know about anybody else, but I am frantically studying the day before, frantically cramming as much stuff into my mind as possible right before the test (simple memorization type stuff), and work as fast as possible the entire time. I have no idea why anybody would plan going to the bathroom (say this is 5 minutes: you just sacrified 1.5-2 points on the exam potentially), or need to; would you rather have a full bladder for an hour or would you rather study 300 hours and lose thousands of dollars of salary? Kinda beating a dead horse to say this to you since you already agree at this point, but I don't know why anybody does it...

yeah, i also can't fathom people going to the bathroom during these exams. every minute is precious. i almost think they have to be cheating in there, because otherwise they are just shooting themselves in the foot.

darkfairy
05-10-2011, 09:00 AM
I went to the bathroom twice during the exam..

Since I started reserving part first, I thought it was tolerable. Didn't expect the pricing part can take sooooooooooooooooooooooo long....



yeah, i also can't fathom people going to the bathroom during these exams. every minute is precious. i almost think they have to be cheating in there, because otherwise they are just shooting themselves in the foot.

Vorian Atreides
05-10-2011, 09:20 AM
Wasn't there a sign out sheet for the restrooms at your facility?

booyah81
05-10-2011, 10:16 AM
yeah, i also can't fathom people going to the bathroom during these exams. every minute is precious. i almost think they have to be cheating in there, because otherwise they are just shooting themselves in the foot.

i spoke to someone at my exam site that said the CAS has someone make rounds through the bathrooms before and during each exam (he was "the guy" for our site), and that he heard they removed a cheat sheet from a chicago site. if anyone knows someone who failed an exam miserably in chicago... ;-)

Bored
05-10-2011, 11:02 AM
yeah, i also can't fathom people going to the bathroom during these exams. every minute is precious. i almost think they have to be cheating in there, because otherwise they are just shooting themselves in the foot.

Wow, are you people not really able to fathom that some people have smaller bladders than you and are physically incapable of going more than 4 hours without a bathroom break?

JasonScandopolous
05-10-2011, 12:22 PM
Wow, are you people not really able to fathom that some people have smaller bladders than you and are physically incapable of going more than 4 hours without a bathroom break?

Unless they have a diagnosed medical problem and/or are on medication, nope, not able to fathom.

Scarlett5
05-10-2011, 12:39 PM
I took an exam in 2006 8 and a half months pregnant and still didn't use the bathroom :) Of course I didn't drink anything all morning in order to make that possible.

Bored
05-10-2011, 01:13 PM
Unless they have a diagnosed medical problem and/or are on medication, nope, not able to fathom.

But it's easier to fathom that the people going to the bathroom are just not bright enough to realize it takes time away from the test. Gotcha.

JasonScandopolous
05-10-2011, 01:40 PM
But it's easier to fathom that the people going to the bathroom are just not bright enough to realize it takes time away from the test. Gotcha.

Yes, it is easier to fathom that people don't realize how valuable time on exams is and that 90+% of people who go to the bathroom during exams did not actually have to do so. e.g. 8.5 month pregnant woman above didn't go, what is anybody else's excuse absent a medical problem?

I'm not saying that people don't go without it having any consequences; but, these are the people that studied for a 10 and are being arrogant in terms of their predicted ability to finish and do well. On the other hand, if you went to the bathroom and didn't finish the test, you screwed up and you should have done something differently that morning so that you didn't.

ThePerfectGame
05-10-2011, 02:53 PM
Adult diapers run about $14 for an 18-pack. A one-time investment and you're covered for 2 sittings per exam. Or, if you're confident, you could go halfsies with a similar-sized buddy for just 7 bucks. Heck, you'll spend more on the 3 bottles of water you're now able to consume worry-free. Just sayin'...

Bored
05-10-2011, 03:09 PM
8.5 month pregnant woman above didn't go, what is anybody else's excuse absent a medical problem?


Uh, a smaller bladder, as I said before? You don't have to have an actual medical condition to have that. I've been pregnant, and I didn't have every stereotypical problem. Just because this woman didn't have have to go to the bathroom often doesn't mean nobody else ever had that urge more than her.

I'm giving up on you ever understanding anything outside your own experience. It's just crazy to me that you think people can study this hard and just throw it all away without thinking about the time being wasted unless they absolutely had to.

Gyudon
05-10-2011, 04:02 PM
Here's a groundbreaking idea:

Have a 1-2 minute break halfway through.

TRINIDON2K
05-10-2011, 05:09 PM
you don't lose points for pissing yourself...

I needed to go to the bathroom bad during the exam but I didn't want to lose time... As soon as they said we were dismissed I was basically throwing people out of the way to get to the bathroom... hahaha

A lot of people went to the bathroom during the test at my sitting... It's my guess that anyone who did, probably failed.

Sssuperdave
05-10-2011, 05:34 PM
A lot of people went to the bathroom during the test at my sitting... It's my guess that anyone who did, probably failed.

This is probably TMI, but I've used the bathroom once during every 4-hour exam I've taken. I don't think it has hurt me at all, as I'm usually thinking about how to approach a difficult problem, and sometimes have an insight that helps me solve the problem. So far, I'm 3 for 4 on those exams.

Kongo
05-10-2011, 06:21 PM
someone should send the last 10 posts or so to the CAS Exam Comittee

kyleucf
05-10-2011, 09:16 PM
I use the restroom every test. I just cant think if I gotta go. Takes under 2 mins... Plus the boost that the coffe gives me outways the time to...ya know....

jjb3113
05-10-2011, 10:29 PM
Geez, are we really talking about bathroom breaks? I routinely have gone twice for 4 hour exams. I find the time is not wasted, but allowing my mind and hand a break. I am over half the way through the exams so do not think it has hurt me. I even took 2 breaks when I passed Exam C at prometric and anyone taking the CBT exam knows how painful the sign-out process is with them.

CASualty
05-11-2011, 10:39 AM
Adult diapers run about $14 for an 18-pack. A one-time investment and you're covered for 2 sittings per exam. Or, if you're confident, you could go halfsies with a similar-sized buddy for just 7 bucks. Heck, you'll spend more on the 3 bottles of water you're now able to consume worry-free. Just sayin'...

I've heard of people writing the bar exam (or something similar) and actually wearing diapers. Apparently the room starts to smell like pee about halfway through the exam.

CASualty
05-11-2011, 10:42 AM
you don't lose points for pissing yourself...

I needed to go to the bathroom bad during the exam but I didn't want to lose time... As soon as they said we were dismissed I was basically throwing people out of the way to get to the bathroom... hahaha

A lot of people went to the bathroom during the test at my sitting... It's my guess that anyone who did, probably failed.

I never go to the bathroom during exams (no coffee in the morning) but I can understand that some people may need to. IMO, it is better to give up two minutes to go to the bathroom than to feel like you have to go for an hour. I wouldn't be an efficient exam writer if I feel like I have to pee REALLY bad!

doop
05-11-2011, 11:24 AM
First chance to get back to AO. Some thoughts:

(1) I did not finish, so I'm somewhat relieved to see that's a common theme. I left 6.25 pts blank, and having looked at the exam again I know I got some other ones wrong too.

(2) Even though I didn't finish, I felt the exam was a good length. I came in feeling underprepared (only had 2.5 months to study, as I took C/4 in Feb), so I expected to have trouble finishing. For those complaining about calculations, I definitely understand. However, if you know the material well enough, you should be able to see a problem and almost instantly know how to do it. If you ran out of time, it wasn't only due to the number of calculations, but also the time it took you (us) to figure out which calculations to perform. A truly qualified candidate wouldn't take as long as I did to figure out the questions, because a qualified candidate would know how to do every question that's ever been asked. The exam is supposed to be hard, especially if you don't know the material well! If you ran out of time, you simply didn't know it well enough, end of story.

(3) Who cares about this bathroom thing? If someone needs a break, that's fine, and if 2 minutes is the difference between a pass and a fail, see (2) above.


That's all I got. Like I said I did not finish, and I'm usually fast (I left C an hour early and got a 10). From what I see here, don't worry too much if you didn't finish. It's all relative and this exam has never been given before, so I'd be very surprised if they firmly stick to the "a priori" pass mark if the actual scores are consistently low. But what do I know?

Best of luck to everyone!!

Fuzzy13
05-11-2011, 11:40 AM
I heard about this study a while back, this would for sure be useful for all the 1 pointers:

http://www.psychologytoday.com/blog/choke/201104/needing-pee-enhances-decision-making-really

Vorian Atreides
05-11-2011, 11:47 AM
I heard about this study a while back, this would for sure be useful for all the 1 pointers:

http://www.psychologytoday.com/blog/choke/201104/needing-pee-enhances-decision-making-really
Note that the "test" was based on readily ingrained knowledge--do you know your colors?--and concludes that focus is sharpened, not recall.

I'll bet that when recall of "unfamiliar" knowledge (that which isn't used "everyday") is actually involved, performance will start to deteriorate.

oblivious
05-11-2011, 12:43 PM
So why is the exam not posted yet?

It usually is by this time?

Irish Blues
05-13-2011, 12:54 PM
I didn't see this mentioned in the thread, but did anyone get a packet labeled "ISO Personal Automobile Manual" that was instead a copy of the exam?

chopsuey
05-13-2011, 01:15 PM
Yes, it was a blue copy of the exam. Didn't realize it until late into the exam.

Irish Blues
05-13-2011, 02:43 PM
According to the version I heard, the CAS did send out a corrected version the morning of the exam ... via e-mail at about 8:15am Central. Unfortunately, candidates had to be in the exam room by 8:00am - which means at a minimum, proctors in the Eastern and Central Time Zones didn't know anything about it until after the exam was completed.

So glad I knocked this exam out 2 years ago.

MightySchoop
05-13-2011, 02:49 PM
I didn't see this mentioned in the thread, but did anyone get a packet labeled "ISO Personal Automobile Manual" that was instead a copy of the exam?

Was this an issue with 5A or full 5?
We had 6 candidates for full 5, but none for 5A, and we had no problems at our exam site.

Vorian Atreides
05-13-2011, 02:55 PM
Was this an issue with 5A or full 5?
We had 6 candidates for full 5, but none for 5A, and we had no problems at our exam site.
Both . . . those not in Central/Eastern Time Zones are likely to not have any issues since the proctors had time to make the necessary changes in advance of the Exam time.

FourKicks
05-13-2011, 02:57 PM
Was this an issue with 5A or full 5?
We had 6 candidates for full 5, but none for 5A, and we had no problems at our exam site.

it was strictly a 5A issue.

i mentioned in the 'ISO PAM' thread in the Exam 5A subforum why i believe question #1 has to be tossed out or treated as a bonus question.

Bobby
05-15-2011, 03:06 PM
I had a dream last night that the pass list came out and my name wasn't on it. But I was on some other list called "Just barely failed". I hated that list.

snoo
05-15-2011, 03:31 PM
I had a dream last night that the pass list came out and my name wasn't on it. But I was on some other list called "Just barely failed". I hated that list.

Did you see my name on that list? It sure wouldnt surprise me.

Michael D Wang
05-15-2011, 11:37 PM
I didn't see this mentioned in the thread, but did anyone get a packet labeled "ISO Personal Automobile Manual" that was instead a copy of the exam?

seems no such issue in Beijing though we took the exam around 12 hours earlier than EST....:lol:

Guest
05-16-2011, 09:28 AM
According to the version I heard, the CAS did send out a corrected version the morning of the exam ... via e-mail at about 8:15am Central. Unfortunately, candidates had to be in the exam room by 8:00am - which means at a minimum, proctors in the Eastern and Central Time Zones didn't know anything about it until after the exam was completed.

So glad I knocked this exam out 2 years ago.

Wow. The proctor at my exam did not stop candidates from reading the PAM prior to the reading period. This is when numbering exam answers and reading exam instructions which I thought was screwed up but this is worse.

shab
06-08-2011, 08:22 AM
when is the next exam 5 sitting ?

JasonScandopolous
06-08-2011, 08:28 AM
when is the next exam 5 sitting ?

almost certainly early May 2012.

shab
06-08-2011, 08:45 AM
almost certainly early May 2012.

Thats crazy.. i thought it will be in november 2011.

JasonScandopolous
06-08-2011, 09:08 AM
Thats crazy.. i thought it will be in november 2011.

all of the upper level exams are offered once per year, only... gotta pass these ones the first time around if at all possible...

nonactuarialactuary
06-08-2011, 09:50 AM
Thats crazy.. i thought it will be in november 2011.

In the new system, odd exams (5, 7, and 9) will be offered in early May, and even exams (6 and 8) will be offered in the late October.

Be glad you didn't sit for old exam 7 in May 2010. Anyone who failed that exam has 18 months in between sittings due to the transition. The nation-specific accounting exam won't be offered again until October 2011 as new exam 6.

shab
06-08-2011, 10:54 AM
Thanks for the info guys.
@nonactuarialactuary, i see you have completed online course 1, how long did it take and whats the level of difficulty and time spent on studying?

nonactuarialactuary
06-08-2011, 01:24 PM
Thanks for the info guys.
@nonactuarialactuary, i see you have completed online course 1, how long did it take and whats the level of difficulty and time spent on studying?

I didn't actually take it. I have credit for it from passing exam 5 in the old system. I'm taking Online Course 2 in July though so I'll let you know how that one goes. So far, everything is pretty straightforward, but I've seen the majority of the material before on old 5 and old 6.

shab
06-10-2011, 12:08 PM
I didn't actually take it. I have credit for it from passing exam 5 in the old system. I'm taking Online Course 2 in July though so I'll let you know how that one goes. So far, everything is pretty straightforward, but I've seen the majority of the material before on old 5 and old 6.

Great, thanks,,