So what did you all think?

[Chance123]

I thought the morning session went ok. Afternoon not as good.

I couldn't work out the CVAT/MEC question.

And did anyone memorize the formulas needed to do the GMMB calculations???

[neustadter]

I thought the morning session went ok. Afternoon not as good.

I couldn't work out the CVAT/MEC question.

And did anyone memorize the formulas needed to do the GMMB calculations???

I couldn't do this problem since I don't know commutation functions. Were commutation functions on the sylabus? I know that some of the readings used commutation functions, but didn't go into detail.

I've never had commutation functions on any of my exams. Has anyone else not had commutation functions?

[Roto]

I thought if one article was going to be heavily tested it would have been Macro-Pricing... not the stupid World Class PD article. I thought those Q's were pretty bad... but then I've been *****ing about that article since day 1. I just don't find it relevant, and clearly was not able to pull out the "important" (testable) material.

I also was pretty surprised at the lack of LIPF material.

Overall I was pretty happy with my performance in the sense I think given what I "knew", I think I regurgitated what I could... but there are definitely some big gaps. Probably a '5'. Overdue for a '6' though, so :fingerscrossed:

[Roto]

I couldn't do this problem since I don't know commutation functions. Were commutation functions on the sylabus? I know that some of the readings used commutation functions, but didn't go into detail.

I've never had commutation functions on any of my exams. Has anyone else not had commutation functions?

Me too.

I just stated my assumptions and how I was interpretting them, and went from there.

Ran out of time bigtime on this one though, so I'd be happy with 1/5.

[Chance123]

The question that asked us to solve for unknown mortality under 1) preservation of total deaths and 2) conservation of total deaths... aren't they the same concept?

[wat?]

Wat the hell was that.

[PVaultGuy]

Anyone get 10.25 instead of 9.25 for the option budget cost - 95% participation rate. Was this q defective or am I defective?

I couldn't do the Dx Mx Nx. What the heck is Mx and Nx ????

[HLD]

I got the fund available for heging budget is $10,250 i.e. 10.25%. Same as what you got.

[Chance123]

I got the fund available for heging budget is $10,250 i.e. 10.25%. Same as what you got.

I got this as well. Then I assumed that the relationship between option cost and participation rate is linear (it appeared to be so from the 3 set of values they provided) and said we could afford a 100% participation rate.

[P. Tear Griffin]

I got the fund available for heging budget is $10,250 i.e. 10.25%. Same as what you got.

i also got 10.25%

[HLD]

Chance123
That's clever... i haven't thought of it... can't remember the relationship of the option costs though.
I was hoping for more calculation Qs in the afternoon, but couldn't do any of them :-(

[PVaultGuy]

I'm glad I'm not defective. I went ahead and used a 95% participation rate, since I didn't want to assume it was linear beyond what the table gave (.75% between each on the table). Hopefully I'll get a couple points. I'll probably still write the soa and claim the question was defective.


And did anyone memorize the formulas needed to do the GMMB calculations???
I don't recall what the GMMB question was asking, a little hint may jog my memory.

The question that asked us to solve for unknown mortality under 1) preservation of total deaths and 2) conservation of total deaths... aren't they the same concept?
I put the same answer for both. That's static hedging I think.

[Uncle Howie]

The question that asked us to solve for unknown mortality under 1) preservation of total deaths and 2) conservation of total deaths... aren't they the same concept?Can someone please address those? I looked at that question and asked myself are these guys retarded? It's the same thing dude. But more likely than not, I'm the one who's retarded and it's not the same.

[Logan 5]

Anyone get 10.25 instead of 9.25 for the option budget cost - 95% participation rate. Was this q defective or am I defective?

I couldn't do the Dx Mx Nx. What the heck is Mx and Nx ????

Mx is the level-benefit commutation function and Nx is the level-premium commutation function. They all relate by the formula:

Mx = Dx - d*Nx, where d = i/1+i

The key thing, as shown somewhere in LIME, is that Ax = Mx/Dx, and ax = Nx/Dx, which are the two values you need to do CVAT and MEC.

[Logan 5]

And why couldn't the Hardy Ch 9 question just give a list of simulated losses instead of requiring exact calculation from a lognormal distribution?

The problem was long enough as it was with all the other parts. :swear:

[wat?]

Uh ... crap. That's a lot of stuff you guys are discussing.

I had the same thing for the selective mortality thing - "AAAAAH aren't these the same?" Dumb or not - I wrote verbatim "This is the calculation for either i) or ii) - I'm not sure which one."

I bombed the Index Account Value question (13 pts, wth) - I couldn't figure out how to pick the participation rate. So, I wrote "I'm just going to use 90% - I have no justification for my choice, I'd just like to demonstrate the mechanics of the IAV."

Almost completely blanked on the GMWB problem, and couldn't figure out how to do the calculations for the LI/MEC problem. Completely whiffed on the lognormal calculation (ended up assuming the QRM = 0.90, which was 1.28 standard deviations away and the CTE = 0.953, so it was 1.67 standard deviations away).

So, in short, got raped by the calculation problems, did less than mediocre on the list/description problems.

Man. Well, I'll look forward to starting the DP-ILA thread for 2008!

[RichieGB]

The key thing, as shown somewhere in LIME, is that Ax = Mx/Dx, and ax = Nx/Dx, which are the two values you need to do CVAT and MEC.

Wasn't it a 20-year term product, though? That means Ax should have been (M_x - M_(x+20)) / Dx, and ax = (Nx - N_(x+7)) / Dx (for MEC premium)...but I couldn't get anything to work out in the problem.

[RichieGB]

I got this as well. Then I assumed that the relationship between option cost and participation rate is linear (it appeared to be so from the 3 set of values they provided) and said we could afford a 100% participation rate.

This is exactly what I did. I swore it was defective until it dawned on me that they're probably just being tricky!

[mjb]

Uh ... crap. That's a lot of stuff you guys are discussing.

I had the same thing for the selective mortality thing - "AAAAAH aren't these the same?" Dumb or not - I wrote verbatim "This is the calculation for either i) or ii) - I'm not sure which one."

I bombed the Index Account Value question (13 pts, wth) - I couldn't figure out how to pick the participation rate. So, I wrote "I'm just going to use 90% - I have no justification for my choice, I'd just like to demonstrate the mechanics of the IAV."

Almost completely blanked on the GMWB problem, and couldn't figure out how to do the calculations for the LI/MEC problem. Completely whiffed on the lognormal calculation (ended up assuming the QRM = 0.90, which was 1.28 standard deviations away and the CTE = 0.953, so it was 1.67 standard deviations away).

So, in short, got raped by the calculation problems, did less than mediocre on the list/description problems.

Man. Well, I'll look forward to starting the DP-ILA thread for 2008!

You read my mind on a lot of these questions. I did exactly the same thing for the conservation of deaths questions. I really don't recall two different calculations. I also picked 90% for the participation rate in the EIA problem since I couldn't figure out how to get the profit.

No clue about the lognormal calculations. I was so prepared for quantile and CTE calcs using stochastic simulations.:redcard:

Roto, I totally agree about the "World Place Prod. Dev." questions. I read this note but specifically ignored it because of all the notes, I thought this would be the last one they'd test. How could they not test Macro pricing!!

It also had a disproportional amount from the of the smaller, less-important (at least my opinion) study notes. Why have a whole question on ending the mortality table when the note basically said it doesn't matter how you do it?

Overall I thought the test was pretty bad. Didn't test important things I thought for sure it would and had some pretty unclear questions. I understood what they were asking but I didn't think it corresponded to any list I had seen. I as able to answer but I have no idea if it will match up to what they're looking for. That said, I was able to put down what I thought were reasonable answers for most questions so I don't feel too bad about my performance.

For me, the morning was much worse than the afternoon. I was feeling pretty discouraged at lunch time.

[P. Tear Griffin]

there were some that i got 100% of if i was putting the right list, and 0 otherwise.

what list was intended for the pricing assumptions question on the UL with a no lapse guarantee and 10th year bonus (i believe it was #17)?

what were the 3 main commission scales?

[100in1998]

Anyone get 10.25 instead of 9.25 for the option budget cost - 95% participation rate. Was this q defective or am I defective?

I couldn't do the Dx Mx Nx. What the heck is Mx and Nx ????

I got something around that. I just stated that the 95% option budget would fall within the constraints.

[WRXSTi]

The afternoon part of that exam took my anal virginity and my dignity along with it.

Why the **** was there a 7702 question regarding a ROP term product, and why the **** did they use commutation functions? I'm not from the stoneage, don't give me that bullshit.

What is with that GMMB problem? I was expecting something where you calculated hedge error or something?

LTCI?!? That was like two ****ing pages of the syllabus and they made an 11 point problem.

WHAT THE **** IS A "FUZZY FRONT END"?? That question made me so pissed, I don't remember reading that at all.

[WRXSTi]

Is it just me or did they pick out some reaaaally abstract problems that covered a very small portion of the syllabus?

[WRXSTi]

BTW...I got 10.25% on the option budget as well. I said this allowed for us to use the highest part rate listed because it was still under the option budget.

[100in1998]

there were some that i got 100% of if i was putting the right list, and 0 otherwise.

what list was intended for the pricing assumptions question on the UL with a no lapse guarantee and 10th year bonus (i believe it was #17)?

what were the 3 main commission scales?

I'm not sure about the pricing assumptions, so I listed a general assumption building list (use actual or similar data, structure of assumptions, etc).

Comm Scales: Level, Levelized, and Heaped.

[inexactuary]

The question that asked us to solve for unknown mortality under 1) preservation of total deaths and 2) conservation of total deaths... aren't they the same concept?

One assumes that non-selective lapses have the same mortality as the persisting group, and the other assumes that they have normal mortality. The first one is from the S/U ART article, but I have no idea which one is named which. What a retarded question.

[inexactuary]

I'm not sure about the pricing assumptions, so I listed a general assumption building list (use actual or similar data, structure of assumptions, etc).


I did the same and tried to tie it to the specific design.

[inexactuary]

Is it just me or did they pick out some reaaaally abstract problems that covered a very small portion of the syllabus?

Yep, they definitely hit the published references pretty hard.

[WRXSTi]

One assumes that non-selective lapses have the same mortality as the persisting group, and the other assumes that they have normal mortality. The first one is from the S/U ART article, but I have no idea which one is named which. What a retarded question.

Interesting. I used the S/U ART formula for the COD and used the formula that looks like:

((1-Qwnorm-Qwnon-sel)*Qdnonsel - Qdsel*Qwsel)/(1-qwnorm-qwextra)

for the preservation of deaths. Probably wrong but I put a lot of garbage on that exam hoping for partial credit.

[inexactuary]

WHAT THE **** IS A "FUZZY FRONT END"?? That question made me so pissed, I don't remember reading that at all.

I don't remember this exact terminology. They could have made it up for all I know, but I guessed that they were talking about the opposite of concurrent engineering, where the front end and back end aren't well defined in the process.

[rekrap]

what were the 3 main commission scales?

They are addressed in a blue box in LIPF. Suggestion for next time, if you unfortunately wind up having to take DP again, is to review the 30-40 blue box examples in the rquired chapters, as at least one of them will be on the exam... lazy test writers. :judge:

[rekrap]

Yep, they definitely hit the published references pretty hard.

And they mixed and matched them strangely... LTCI with Four-Fields? Inflation-Indexed annuities with Ending the mortality table? :crazy:

[RichieGB]

And they mixed and matched them strangely... LTCI with Four-Fields? Inflation-Indexed annuities with Ending the mortality table? :crazy:

Did other people take a page from model solutions to old problems? For instance, when it asked about Four-Fields in the context of the LTCI problem, I just talked about Four-Fields and didn't relate it to LTCI. Same with asking about the mortality table within the context of II Annuities.

Hope the grading style hasn't changed.

[rekrap]

Did anyone else feel that 5(b) and 6(a)(ii) were basically the same question about the feasability stage of the product design process?

Double whammy.

[inexactuary]

Did anyone else feel that 5(b) and 6(a)(ii) were basically the same question about the feasability stage of the product design process?

Double whammy.

Yes. Feasability shows up in more than one place on the syllabus, and my theory is that the two writer's were referring to the two places and one hand didn't know what the other was doing. I remember that one list was specifically phrased in terms of questions to asked, which follows the wording of the 6 question. I just memorized one macro list, and they are both basically the same, so you should get credit for both. The second one asked for relevance also, so you had to tie it to the specific situation.

[inexactuary]

Did other people take a page from model solutions to old problems? For instance, when it asked about Four-Fields in the context of the LTCI problem, I just talked about Four-Fields and didn't relate it to LTCI. Same with asking about the mortality table within the context of II Annuities.

Hope the grading style hasn't changed.

I tried to relate to the specific situation here and there, but for the most part just stuck to the "script"

[rekrap]

Yes. Feasability shows up in more than one place on the syllabus, and my theory is that the two writer's were referring to the two places and one hand didn't know what the other was doing. I remember that one list was specifically phrased in terms of questions to asked, which follows the wording of the 6 question. I just memorized one macro list, and they are both basically the same, so you should get credit for both. The second one asked for relevance also, so you had to tie it to the specific situation.

There is a committee reviewing all the questions, and they should have caught it.

Was there a reference in one to a particular study note to use in answering the question? I think there wasn't, making a "generalized" answer acceptable for both, even if it wasn't intended. But still....

[inexactuary]

There is a committee reviewing all the questions, and they should have caught it.


Well, they also thought distinguishing between preservation and conservation of deaths was a good question, so they're obviously not perfect.

[rekrap]

Well, they also thought distinguishing between preservation and conservation of deaths was a good question, so they're obviously not perfect.

IMHO, they also got cute with the four fields map, the CVAT/MEC commutation functions and the exact calculation of risk measures.
This isn't a time to be cute or clever, it's a time to assess our understanding of the concepts.

[RichieGB]

IMHO, they also got cute with the four fields map, the CVAT/MEC commutation functions and the exact calculation of risk measures.
This isn't a time to be cute or clever, it's a time to assess our understanding of the concepts.

That was very frustrating re: commutation functions. I hope that the seminar instructors are right in the sense that setting up and explaining the ideas behind the CVAT and MEC test should be enough to get most credit.

Incidentally, for that calculation, my solution to (i) was probably wrong given the wording of (ii), and then my solution to (ii) was probably wrong given the wording of (iii).

[Britney]

Does anyone have any idea whatsoever when results are anticipated to be released? Is it 6-8 weeks for this one like the preliminary exams or is it more like 8-10?

Thanks!

[wat?]

I'm not sure about the pricing assumptions, so I listed a general assumption building list (use actual or similar data, structure of assumptions, etc).

Comm Scales: Level, Levelized, and Heaped.

Does anyone have any idea whatsoever when results are anticipated to be released? Is it 6-8 weeks for this one like the preliminary exams or is it more like 8-10?

Thanks!

Expect it sometime the 2nd or 3rd Friday of January. So, 10-11 weeks.

[Actuary_JB]

It sounds like the exam had some crazy questions on it! It's always hard to know what to expect going in to a "new" exam, so I'm guessing most people were surprised by some of the questions that appeared (i.e. commutation functions).

Rather than try to address specific questions (since I haven't seen the exam), I'll offer a few thoughts:

1. The really difficult questions will probably only be answered by a few students. These questions just separate the 9's from the 10's. If you can get a few points on those questions, that's great, but don't get too concerned if you feel you "bombed" one or two question.

2. The nice thing about essay exams is that you can receive paritial credit. Thus, if you set up the problem and started it correctly, you can still receive some points even if you got stuck or missed part of the "trick" of the question.

I know it's tough re-hashing the exam, but I hope that you can all enjoy the freedom of not having to study! Have a good weekend.

JB

[Chance123]

The more I read about you guys' answers, the worse I feel about mine. :oops:

[RichieGB]

The more I read about you guys' answers, the worse I feel about mine. :oops:

Don't worry, that's how everybody reading this thread feels. There's probably some psychological term for the phenomenon.

[MWLifer]

The more I read about you guys' answers, the worse I feel about mine. :oops:

Same here. I've gone from "maybe a chance" to filing all my notes, books and readings for next year.... Ugh.

FUZZY? Did that exam have the word FUZZY on it? Twice?!?!?!

[actuariallife]

I felt I didn't have enough energy in the afternoon session, especially during the last hour of the exam.

Anyway, it's gone. Good luck with everyone!

[I Pity the Fool]

And did anyone memorize the formulas needed to do the GMMB calculations???
I did for the Prob(cost of GMMB = 0). It seemed testable and the formula wasn't too difficult. All you really had to memorize were the lognormal parameters and then
1 - Phi[ ( ln (1.05/1.00) - Mu ) / Sigma ]

Had no idea how to do the actual quantile and CTE calculations though.

It sounds like I may have lucked out, if not many people got this one. Those are the kinds of problems I hope to get right, to cancel out the ones I'm getting wrong that everyone else got right.

Study hard in school!!

[inexactuary]

I did for the Prob(cost of GMMB = 0). It seemed testable and the formula wasn't too difficult. All you really had to memorize were the lognormal parameters and then
1 - Phi[ ( ln (1.05/1.00) - Mu ) / Sigma ]


Don't forget the management fee.

[rekrap]

Don't forget the management fee.

And the five years...?

[100in1998]

I'm switching to the CSP thread. I can't keep re-hashing this exam. It will drive me crazy! :crazy:

Hope to see some of you over there! (BTW, the syllabus is out)

[PVaultGuy]

FUZZY? Did that exam have the word FUZZY on it? Twice?!?!?!
I said that "this is inappropiate to discuss in this setting" refering to "Fuzzy front-end". Does anyone know what source this is from?

[inexactuary]

Does anyone know what source this is from?
World Class Product Development study note, pg 45-46

[MWLifer]

I said that "this is inappropiate to discuss in this setting" refering to "Fuzzy front-end". Does anyone know what source this is from?

Playboy - Jenny McCarthy Issue circa 1994

[rekrap]

I'm switching to the CSP thread. I can't keep re-hashing this exam. It will drive me crazy! :crazy:

Hope to see some of you over there! (BTW, the syllabus is out)

Just wait until they release it. You'll be back soon. :)

[I Pity the Fool]

Don't forget the management fee.And the five years...?
They're incoporated as part of the mu and sigma parameters, or at least they were in my formula. There's no guarantee I did it right, but it did give me one of the five available values in the normal table (-1.19?).

Mu = n * (u + ln(1-m))
Sigma = n * (sigma^2)

I said that "this is inappropiate to discuss in this setting" refering to "Fuzzy front-end".
I went with a reference to a goatee. For "fuzzy after-launch," I put its what you get after a dinner at Taco Bell. :guitar:

[wat?]

I went with a reference to a goatee. For "fuzzy after-launch," I put its what you get after a dinner at Taco Bell. :guitar:

Fuzzy? What'd you do, eat the chihuahua too? :)

[wat?]

The more I read about you guys' answers, the worse I feel about mine. :oops:

I TOTALLY AGREE. I mean, I'd feel bad about your answers, too.

:) I keed, I keed.

Don't worry, that's how everybody reading this thread feels. There's probably some psychological term for the phenomenon.

I used to call it "I'm an idiot". There's probably a better technical term for it. Like, "fuzzy brained" or something.

[rekrap]

I used to call it "I'm an idiot". There's probably a better technical term for it. Like, "fuzzy brained" or something.

"Fuzzy front-end" or "Fuzzy back-end"?

[Chubbs_23]

- I also got 10.25% for the option budget and chose to use the 95% part. rate.

- Did they think that 7702 was such an easy subject, that they could only adequately test it if they through it at us with a term + ROP rider, and a table full of undefined commutation functions? Or with an LTCI rider on a UL policy?

- Strange that there were no real questions on reinsurance, non-forfeiture law, or required capital... but two questions on phrases including the word "fuzzy".

- When #16 asked for "immunization" techniques, I think that is limited to cash flow matching, duration/convexity matching, and a combination of that. Anything else would fall under ALM techniques IMO. Sound right?

-#17(a) asked about pricing assumptions for flexible premium UL product with a lifetime NLG. I think the phrasing of that question was extremely vague, and I'm not sure what they were looking for as a model answer.

[Chance123]

I TOTALLY AGREE. I mean, I'd feel bad about your answers, too.

:) I keed, I keed.

Thanks for the love wat.

[rekrap]

- Did they think that 7702 was such an easy subject, that they could only adequately test it if they through it at us with a term + ROP rider, and a table full of undefined commutation functions? Or with an LTCI rider on a UL policy?

See my earlier post about them trying to be clever instead of just assessing.


- When #16 asked for "immunization" techniques, I think that is limited to cash flow matching, duration/convexity matching, and a combination of that. Anything else would fall under ALM techniques IMO. Sound right?
Yes. Exact matching, Duration matching and horizon matching is what I hope they were expecting. :tup:


-#17(a) asked about pricing assumptions for flexible premium UL product with a lifetime NLG. I think the phrasing of that question was extremely vague, and I'm not sure what they were looking for as a model answer.
I just spit out the Experience SN 107 list. And then tried to suggest key ones to develop in step one (obligation, asset, scenario) specific to the product.

[P. Tear Griffin]

i think (hope) mentioning and explaining exact matching, duration and convexity matching, and horizon matching would get full credit. i am pretty sure that is what they were looking for.

for #17 (a) i put the "steps in developing experience assumptions" list. is that SN 107?

a few other exam questions i can remember:

the speculative grade corporate bond question- was there a source that recommended a speculative grade bond strategy, or was this purely an individual recommendation?

the preservation/conservation of total deaths- like many people, i thought these were the same thing. hopefully my answer was right for 1 of the 2. i think it was between 5.4 and 5.5.

the profit margin question immediately following the preservation of total deaths part, where something like 5% of preferred risks were actually non-preferred- i think the original profit margin was 4.1%

types of variable income annuities and how effective they are at inflation protection- i put a fixed % COLA, increase tied to CPI, and increase/decrease by factor of NIF/AIR

[wat?]

the speculative grade corporate bond question- was there a source that recommended a speculative grade bond strategy, or was this purely an individual recommendation?

I just suggested that they cover the guarantees in the SPDA (if any) with "risk-free bonds" (which, in retrospect, is likely incorrect - should have used 'bonds at current rates') and invest a portion of surplus in spectulative-grade bond strategy.

I asserted that risky assets should not be used to back guarantees and that one should consider that increasing the investment in risky assets leads to increasing the asset default factor of risk-based capital, so you don't want to invest in TOO much speculative-grade bonds.

[leathercost]

Did anyone have idea about the additional cost of complexity for GMWB?

[inexactuary]

I spit out the lists from the Hidden Costs of Product Complexity/Hidden Costs of Complex Products articles.

[wat?]

Did anyone have idea about the additional cost of complexity for GMWB?

I spit out the lists from the Hidden Costs of Product Complexity/Hidden Costs of Complex Products articles.

AAAAAAAAAAAAAAAAAH dammit

[cat20]

Well I thought I had nailed the commutation formulas but now I realize I forgot the product was a 20 yr term. Oh well. Still hoping for an A for effort. :roll:

[StephonMarbury2]

Am I the only one who recommended a participation of greater than 100% based on the linear nature of the participation rates vs. option budgets and an option budget of 10.25%?

I was a little disappointed that the exam focused more on 'memorization' and less, I felt, on understanding/application. I guess at some point they are the same but there were a lot of obscure stand-alone questions.

I also cannot believe they asked the exact calc of the quantile/CTE. What a disaster trying to answer that one.

[Chubbs_23]

I just spit out the Experience SN 107 list. And then tried to suggest key ones to develop in step one (obligation, asset, scenario) specific to the product.

I'll be interested to know the result of this one. I intentionally avoided answering from that SN because they didn't use the word "experience" in the question. However, if that is the answer I'll feel stupid, because I knew that SN word for word!

As for the speculative grade bond question, I talked a bit about speculative bonds from the Moodys standpoint, addressed the impact of default costs on the net investment rate. I recommended a small exposure to them, to help increase yield.

Another interesting question was where they asked about "pricing pitfalls" of the EIA. There is the popular LIPF list on "pricing pitfalls", but there is also another study note that talks about common pricing mistakes in equity linked products (macro product management, savvy buyers, short sighted hedging, seeking perfect hedge...) IF they wanted stuff from the latter study note which is more geared towards EIA's, I think its very misleading that they used the word "pitfalls" in the question.

[RG]

Thanks for all your comments. I felt AWFUL walking out of this exam and it does make me feel better to know that I wasn't the only one. There was only one other person taking this exam at my sitting and that person left half way through the afternoon.

My thoughts are:
1. I used to teach college calc and I had a habit of giving exams like this, which mostly focussed on separating the best from the okay. I'm not sure what the point is of that here, since a pass is a pass, but it will scare future students into learning the material better.

2. I really wish I had had more calculation problems to practice on. Knowing the theory isn't enough for me in the face of exam panic. I almost spaced out on the formula for modified duration even though we did those problems for the FAP; I kept putting down convexity and then saying "no, that's convexity". My concentration bit the dust around an hour from the end.

3. I really don't understand what we're supposed to write and what not write. In the middle of the afternoon session I realized that I should have listed the QABs. But the question didn't explicitly ask me to list them. So, should I have? I did not feel like I demonstrated how much I knew - or maybe I did, and that's the problem.

4. I told myself if the DP exam didn't go well, then I was going to start on the CSP right away. Memorized the first 20 cards last night, another 20 today.

5. At one point I thought Bruce had indicated that they wanted the pass rates for DP and CSP combined to be the same as the pass rate for course 8. Given that CSP is the harder exam (and had a pass rate of 50%), I'm hoping that means that there will be a higher pass rate for DP. 75% effective would probably put me in the ballpark.

[wat?]

- When #16 asked for "immunization" techniques, I think that is limited to cash flow matching, duration/convexity matching, and a combination of that. Anything else would fall under ALM techniques IMO. Sound right?

I thought to myself, "THIS must be the reason I had to take Course 6 - just so I could answer this question." I agree with both you and rekrap - the combination of the two was "hybrid matching". I also yammered on with the %age price change = -Duration * (change in i) + (1/2) * (Convexity) * (change in i)^2 formula, the weaknesses and strengths of each of the three, and finally calculated the two durations. Ah, if only that question were worth 40 points.

-#17(a) asked about pricing assumptions for flexible premium UL product with a lifetime NLG. I think the phrasing of that question was extremely vague, and I'm not sure what they were looking for as a model answer.

Similar to other questions on the exam, I looked at it and went, "Are you serious."

I did say that the year 10 interest bonus would make lapse rates spike for people that were contributing a lot of premiums (much more than necessary to take advantage of the interest bonus), and that lapses for the remainder of the period would be pretty low b/c the other cohort were people who were paying minimum premiums to keep the policy inforce - they might not be able to get better rates elsewhere, so they'd rather keep paying the guaranteed premium and be backed by the guaranteed benefits.

[Chubbs_23]

Another thing that was interesting. How did they get a 4 point question out of "managing channel conflict"? I mean, I just went back to that chapter of LOMA, and I think I could write out everything the textbook says on it word for word in 6 minutes. Seemed like that question should have been 2pts at most. Unless somebody is going to tell me that there is a "Channel Conflict Study Note" somewhere that I missed!!!

[wat?]

Another thing that was interesting. How did they get a 4 point question out of "managing channel conflict"? I mean, I just went back to that chapter of LOMA, and I think I could write out everything the textbook says on it word for word in 6 minutes. Seemed like that question should have been 2pts at most. Unless somebody is going to tell me that there is a "Channel Conflict Study Note" somewhere that I missed!!!

What'd you recommend for resolving the conflict? I know it was "avoidance" and "integration", but did you put any specific examples or make a recommendation? I tried to put about 2-3 examples for each.

Avoidance:
just don't offer via internet
offer via internet and use career agency to sell other products that have a greater need for agent interaction like whole life, universal life, etc.
market internet product only where agents are not currently marketing (if agents licensed in Hawaii, use internet to market products to the US mainland, wherever the company is licensed to do business).

Integration:
Internet can be used to research term product, and actual purchase done with agent
Agent (or company) can service policy via internet - quicker service, more efficient

... and I think that was it for that bullet point.

I knew it was 4 points, so I made sure to describe what was meant by "different objectives" or "channel cannibalism". I also mentioned I thought it was an example of vertical conflict - in essence, the company would be taking away sales from the career agency by issuing the product directly. It would result in a decrease in commission costs, but would be an increase in costs by monitoring and administering the software/systems required to issue policies over the internet. It would also create unrest in your career agency. And there's no guarantee that you've actually increased your production by issuing to the same market via the internet: you might not be penetrating the market. You could just be stealing sales from the career agency.

[StephonMarbury2]

I don't remember a specific Channel Conflict question. I also took the IC exam, so maybe it was only on IU. Hope I didn't MISS it, cause I would have nailed it!

For resolving it was likely looking for:
- geog excl.
- set up subs. for new channel
- diff target market for new channel

[mjb]

I don't remember a specific Channel Conflict question. I also took the IC exam, so maybe it was only on IU. Hope I didn't MISS it, cause I would have nailed it!

For resolving it was likely looking for:
- geog excl.
- set up subs. for new channel
- diff target market for new channel

I took IC as well and I don't recall a channel conflict question.

From what I could tell, there we only two obvious Canada-specific questions. The policyholder tax question and the preferred term question.

[P. Tear Griffin]

Did anyone have idea about the additional cost of complexity for GMWB?

I spit out the lists from the Hidden Costs of Product Complexity/Hidden Costs of Complex Products articles.

i hope this is enough because i knew everything on those lists but didnt mention anything GMWB specific

[inexactuary]

5. At one point I thought Bruce had indicated that they wanted the pass rates for DP and CSP combined to be the same as the pass rate for course 8. Given that CSP is the harder exam (and had a pass rate of 50%), I'm hoping that means that there will be a higher pass rate for DP. 75% effective would probably put me in the ballpark.

I doubt it will be higher than 50%

[Roto]

I doubt it will be higher than 50%


I wouldn't be suprised if there were some truth to what RG suggested... though 75% seems high. I was going into these things thinking the fall FSA exam would have a slightly higher pass rate than the spring one.

This impression wasn't based on info from any one source, just a general impression I got from communications with multiple sources (this board, webcasts, FSAs who are active in education design process).

So I'll throw 55% effective pass rate out there as my guess.

[RG]

I wouldn't be suprised if there were some truth to what RG suggested... though 75% seems high. I was going into these things thinking the fall FSA exam would have a slightly higher pass rate than the spring one.

So I'll throw 55% effective pass rate out there as my guess.

Yeah, I'm not suggesting 75% will happen. It was meant to be wishful thinking. And honestly, I expect the first csp and dp will have lower pass rates than when it settles out. I can see reasons why the first csp had a lot of people who just missed 8, but also a lot of people who have never taken an essay exam before and decided to take it when fap-2 wasn't released.

Okay, I'll put my money on 63%

[RG]

Another thing that was interesting. How did they get a 4 point question out of "managing channel conflict"?

This wasn't the only problem like this for me. Often it was like "spit out lists, do an easy calc, have 5 more minutes than I think I should have, wonder what I have forgotten to say, stop worrying about it and move on". I'm forgetting already, but the MEC question was all of 8 points but I think it was the duration matching question was 7. On what planet did that MEC question only take 3 more minutes?

My theory: The 3 minutes per point is a good ballpark, but obviously the harder problems take more like 5 minute per point and the easiest ones might be down to 1.5 minutes per point. That's how it worked when I wrote calc exams, I hope the SOA thinks the same way.

[rekrap]

My theory: The 3 minutes per point is a good ballpark, but obviously the harder problems take more like 5 minute per point and the easiest ones might be down to 1.5 minutes per point. That's how it worked when I wrote calc exams, I hope the SOA thinks the same way.

Definitely. The main corollary of that theory is that if you are spending more than 3 minutes per point on any one question, move on; unless you are close to finishing it, then you may do so while still in the mode/mind-frame of working that particular problem, recognizing that you shouldn't spend too much more time, at the detriment of the other remaining problems.

Another corollary, and the point of the read-through time in my mind, is to decide which problem will most likely take you more than 3 minutes per problem (say the MEC calculation), and save that one for after you have given all the other problems their fair share. Then you are less likely to suffer from missed points (because one of the reasons it would take me longer than 3 minutes per problem is because I'd have to take time to think it through and solve, which is where mistakes are likely to enter in).

[RG]

Definitely. The main corollary of that theory is that if you are spending more than 3 minutes per point on any one question, move on; unless you are close to finishing it, then you may do so while still in the mode/mind-frame of working that particular problem, recognizing that you shouldn't spend too much more time, at the detriment of the other remaining problems.

Another corollary, and the point of the read-through time in my mind, is to decide which problem will most likely take you more than 3 minutes per problem (say the MEC calculation), and save that one for after you have given all the other problems their fair share. Then you are less likely to suffer from missed points (because one of the reasons it would take me longer than 3 minutes per problem is because I'd have to take time to think it through and solve, which is where mistakes are likely to enter in).

I think my point was the opposite of your corollary: that it's okay to spend more time on some problems as long as you know where you can make it up. And I definitely don't save the hardest problem for last, because my brain is fried by then. It's a balance - sometimes it's useful to work on it and then stop and "let your subconscious mind work on it" as a prof of mine used to say - and sometimes it's useful to keep plowing through, and just up the upper limit on how much time you'll give it.

[rekrap]

Definitely. The main corollary of that theory is that if you are spending more than 3 minutes per point on any one question, move on; unless you are close to finishing it, then you may do so while still in the mode/mind-frame of working that particular problem, recognizing that you shouldn't spend too much more time, at the detriment of the other remaining problems.


I think my point was the opposite of your corollary: that it's okay to spend more time on some problems as long as you know where you can make it up. And I definitely don't save the hardest problem for last, because my brain is fried by then. It's a balance - sometimes it's useful to work on it and then stop and "let your subconscious mind work on it" as a prof of mine used to say - and sometimes it's useful to keep plowing through, and just up the upper limit on how much time you'll give it.

The first corollary agrees: keep working/writing if you're going strong, but if you're just trying to get more of anything down, move on.

I guess I see your point about not doing the hard ones last, but I usually find that while spitting out some of those four-six pointers first, the back of my mind is trucking away on the 11-13 pointers, which will come in handy when I get to those. Sometimes, if I get an epiphany, I'll even start on the big one, knowing that I can fill in the finer points of the lists on the smaller ones if time permits.

[Uncle Howie]

For the immunization question, isn't immunization specifically duration/convexity matching? I thought cash flow matching was just called exact matching and hybrid matching was hybrid matching. Since they only asked for immunization I only addressed duration and convexity matching.

Did everyone include cash flow and hybrid matching? If so - sh!t...

[rekrap]

Did everyone include cash flow and hybrid matching? If so - sh!t...

If we did, and we weren't supposed to, kudos to you for answering the question asked. :wink:

[rekrap]

For the immunization question, isn't immunization specifically duration/convexity matching? I thought cash flow matching was just called exact matching and hybrid matching was hybrid matching. Since they only asked for immunization I only addressed duration and convexity matching.

Did everyone include cash flow and hybrid matching? If so - sh!t...

Seriously, there was a study note on ALM which posited Immunization as duration matching (with a convexity refinement, as well as a key rate duration refinement), followed by some emerging methods (Option pricing, dynamic hedging and optimization techniques). But both the SN and LIPF call duration matching "immunization", and not the other techniques as such.

[wat?]

For the immunization question, isn't immunization specifically duration/convexity matching? I thought cash flow matching was just called exact matching and hybrid matching was hybrid matching. Since they only asked for immunization I only addressed duration and convexity matching.

Did everyone include cash flow and hybrid matching? If so - sh!t...

I agree - immunization is almost interchangeable with "duration/convexity matching".

But I thought they talked about "methods to balance interest rate risk", not "immunization", specifically?

I forget.

[Kenny]

My memory comes from Course 6 and general finance knowledge, not specifically from this exam (since I didn't take it) but I have always considered interest rate immunization to be the process of minimizing the risk associated with interest rate movements. I thought horizon and CF matching were inclluded in the C6 definition of immunization.

Here are a couple of definitions I found on the interwebs:

http://www.google.com/search?hl=en&q=define%3A+immunization (see the very last def)
In finance, interest rate immunization is a strategy that insures that a change in interest rates will not affect the value of a portfolio. ...
en.wikipedia.org/wiki/Immunization (finance)

http://www.investopedia.com/terms/i/immunization.asp
A strategy that matches the durations of assets and liabilities, thereby minimizing the impact of interest rates on the net worth.

For example, large banks must protect their current net worth, whereas pension funds have the obligation of payments after a number of years. These institutions are both concerned about protecting the future value of their portfolios and therefore have the problem of dealing with uncertain future interest rates. By using an immunization technique, large institutions can protect (immunize) their firm from exposure to interest rate fluctuations. A perfect immunization strategy establishes a virtually zero-risk profile in which interest rate movements have no impact on the value of a firm.

ETA: plus I'm just a stoopid pension actuary so what do I know :tup:

[Chubbs_23]

Why won't they release this thing already? I haven't got to do a commutation based MEC calculation in days!!! I'm in withdrawl!

[wat?]

Why won't they release this thing already? I haven't got to do a commutation based MEC calculation in days!!! I'm in withdrawl!

That's funny, I've never done a commutation function-based MEC calculation.

[Logan 5]

Why won't they release this thing already? I haven't got to do a commutation based MEC calculation in days!!! I'm in withdrawl!

I don't believe they will release them until all the prelim exams are done, too. So later this week. :popcorn:

[wat?]

I don't believe they will release them until all the prelim exams are done, too. So later this week. :popcorn:

You sure? I think the FSA exams have historically been released 2-3 days after the administration of the exams, regardless of when the preliminary exams are administered.

[Logan 5]

You sure? I think the FSA exams have historically been released 2-3 days after the administration of the exams, regardless of when the preliminary exams are administered.

Well, the last prelim exam was today (FM), and I still don't see the DP exams posted on the SOA site. :popcorn:

[wat?]

Well, the last prelim exam was today (FM), and I still don't see the DP exams posted on the SOA site. :popcorn:

Yep. :bunny:

[MWLifer]

Well, the last prelim exam was today (FM), and I still don't see the DP exams posted on the SOA site. :popcorn:

That's OK....I'm still hungover.

[wat?]

That's OK....I'm still hungover.

:tup: Nicely done.

[wat?]

http://www.soa.org/education/resources/edu-multiple-choice-essay-examinations.aspx

You're welcome!

[rekrap]

Did anyone else feel that 5(b) and 6(a)(ii) were basically the same question about the feasability stage of the product design process?

Double whammy.

I had it flipped a little:
5. (13 points) Your company is launching its first Equity-Indexed Single Premium Deferred Annuity.
(a) Explain issues that should be addressed in the feasibility stage of the product development process.


6. (7 points) ...
(b) (3 points) You have been asked to assess the feasibility of a term product.
(i) List key questions to answer during the product development process.
(ii) Explain the relevance of these key questions to the new product.

[inexactuary]

It actually makes me feel better looking at the test again, since now I remember the stuff I did right (I think) and not just the stuff I screwed up.

[wat?]

So what's the chance that for #12, the IIIA part is only worth like, 1/4 of a point, and the Ending the Mortality Table bit is worth the other 6 3/4 points? :)

[rekrap]

So what's the chance that for #12, the IIIA part is only worth like, 1/4 of a point, and the Ending the Mortality Table bit is worth the other 6 3/4 points? :)

Near zero. It is my understanding that they only delineate the point values when there is a significant (i.e., ~2-3 point) difference in value between the parts.

Notice that #11 marks part a as 1 point, part b as 2 points and part c as 5 points. Also, #6 marks part a as 1 point and parts b & c as 3 points each, yet #5 does not demarcate the 13 point across the 4 parts, meaning they are all worth about 3-4 points (so, probably a, b & d are worth 3 while c is worth 4, or some combination thereof).