Go ahead chum. Question 2 was total deja vu - where did I see that before ? Mack and bootstrapping - very nasty but funny stuff. Annoyed with myself for leaving out important bits on the last part of the first question. Overall it wasn't too bad.
Q2 wasn't deja vu... It's assignment 6 of acted Totally inappropriate! I'm sure that question was a gift for most people...but when I looked at the solution of the assignment I realised that I wrote very different things last part of Q1 was vague...don't see what I can say when I don't have enough info besides making questions and recommending investigations... should claims made basis have ibnr?
Why do you think that was inappropriate ? I think the question is basically fair - obviously you remembered a fair bit of the question or else you wouldnt have been able to find it so quickly...I thought the Mack and Bootstrapping thing was far harder - full descriptions... Claims made basis - I didn't know too much about this one with the exception of what I've seen in work (work for a reinsurer) - the big issue there - or so I wrote was not pure IBNR but IBNER - there are two ways in which this can work : cedents or insureds putting in notifications of claims without reporting full numbers - which is why you see the incurreds increasing dramatically. Options on many contracts for delayed notification or extensions Of course you'll have the usual long settlement delays associated with that class as well. It's my third time to sit this paper... so please God I'll pass.
I got 0 for Bootstrapping and Mack. What about the effect on ICA? Isn't the riskiness of the public liability already allowed for in the estimate of the ICA?In any case FSA will review it and provide ICG??It might also depend on the size of the public liability relative to the whole business?? Anyone??
Yup Vinny I think they're all good points. You could also throw in that your calculations show that the effect on IBNR was quite low - I think it was only $3m or so. On the Mack method I wrote a bit about the Munich Chain ladder method - which is funny - cos he wrote a different paper on reserving ranges - so zero for me on that bit. What did you use for last part of Q1 ? My brain was fried at that stage and I believe that was the weakest part of my paper.
Thoughts then? I thought it was a nasty little paper, but very topical - since when have the Mack Method and Bootstrapping been in the core reading? I'd be interested to see the marking schedule for that particular question as I think you could describe them in a widely differing level of detail. Furthermore I'm not sure if some of the markers will know Mack and Bootstrap well enough to know if a candidate is talking sense or not. On the captive question Ian will be glad to hear that I managed to shoehorn "Captives will potentially be worst hit by Solvency II" into one of my answers. Looks like bringing those copies of Insurance Times along to the tutorial was worthwhile!
My initial thought is "Shouldn't this thread be in the Exams Discussion forum ?". Other than that well done , it sounds like you've handled it pretty well.
I calculated loss ratio, is that right for claims made basis using premiums written?I then commented on the volatility, plus more info on the year 2005, something along these lines. I know I started running out of ideas (there were 18 marks, so we needed to write lots!) and then going back to boostrapping as I could not believe that I would get a 0 there..What did U guys write about the last part for question 1?
I wrote some things about Mack. I said that new ultimate is within predicted range. Might impact ICA in that newly assessed volatility of reserves might be higher I too said that it was not IBNR but IBNER and reopened claims I totally messed up last part of Q1 and all of Q2 besides the adv&dis of captives . I wrote about claims analysis stuff.
For the last part I wrote stuff like need for rate-chang analysis, and further investigation into why the premium almost doubled in 2004 - did anyone else pick that up? I wasn't sure why the premium would increase, so I assumed the company is writing loss-making business. More importantly though, do you guys feel that it is fair to have an assignment question in the exam that is there to test ability to apply higher-level skills and not simply a memory test?
this was my first SA3 sitting, but having looked at years of past papers this must be one of the most straighforward (not easiest, but most clear as to what is being asked). when i'm revising for my resit in sept i'll be kicking myself for not getting this one! Q1, I got 0 for mack & bootstrapping as never worked in reserving. I thought last part was the best bit for me. Only calc that can be done is a LR (lots of assumptions), but commented that this is an indication of performance only, not profitability. Need to know expenses, commission, etc to get profit. Then discussed every investigation i can think of under the sun, esp RI (i had time since i spent little time on earlier parts). Q2, like other comments. Most students i guess would have seen this question before!
I got the bit about the premium increasing - I said that it , combined with the reducing profitability was likely due to company's strategy of increasing market share... what I DIDN't include was the really important bread and butter stuff of proper premium rating and checking to make sure the company is not suffering from adverse selection through movement and claims and exposure analysis ... GGGGAAAAHHHH!!! But that's SA3 - there's so much stuff there that you're bound not to remember it all or put it all down in the heat of battle... so cheer up folks and best of luck to you all... I do think question 2 was/is entirely fair - it's certainly a lot fairer than Mack method - as mentioned above Mack wrote THREE papers and not one - I wrote about the wrong one (in truth I didn't know the relevant one). I'd say that in the course of the last 18 months I've put an easy 1000 hours of work into this exam , many hundreds of questions etc etc - to remember all of them is daft ... it's a capital ,reinsurance and experience rating question - with a nice split into the different lines so that you can write enough.
yep, comments on rate changes, strategy (growth vs profit), insurance cycle, or simply it is bad business.
But you're missing the point here: - Mack and Bootstrapping is something that almost everyone would have struggled with, which seems to be the case, given comments so far. - A few individuals would have advanced knowledge on these topics, but that's the same as saying that those who have priced/reserved for public liability insurance would have held an advantage. - With question 2 the issue is that it is something most students would have seen word-for-word, and it goes against the idea of the SA3 exam - which is an "application" paper? As you may have guessed by now, I was one of those who didn't focus on assignments towards the last few weeks!
Funny paper I'm a bit miffed that Q2 is apparently a copy of an Acted question - not really in the spirit of the exams being a fair test of all candidates abilities! I may take this further with the appropriate bodies as it is not on - and suggests a serious conflict of interest between the profession and the educational body which is not recommended in this post-Morris world. I'd be interested to hear Mr.Senator's thoughts on this (other than hooray I predicted a question for all those people that paid Acted a large fee). Apart from that I guess I was lucky as I had thought there might be a ranges question given it is a hot topic in reserving at the moment. I'm interested in whether people think the actuary should provide a range of results as I think there are various interpretations that could made on that question. I think there was more bookwork than people think, I used some fairly standard "reinsurance needs" and "capital needs" lists on Q2 tailored for the captive. Tim
@herewegoagain I take and took your point - but if you check the assignment X6 you'll see that it's not word-for-word (ok I'm being pedantic now) the same , in the assignment it's 5 years, in the question it's 10 years. Maybe I'm just not grade-A actuary material , but I don't imagine that the vast majority of actuaries always try to come up with new and original thought processes to solve the same problems. for example my suspected weakest part of the paper went on and on about trends in the ULR , against the exposure measure (only premium was available) , expenses , reinsurance , cover and conditions but missed out the glaringly obvious and fundamentally important premium rating accuracy tests and associated selection issues. Genius being 95% perspiration and 5% inspiration and all that... I don't claim to be either a genius or perspiration in any case. EDIT : I was bloody amazed and dissappointed there was not a question on the credit crisis and measuring reinsurance risk - there's a great GIRO paper on it by Shaw and it's been perhaps thee topic in the last 12 months.
I know what you mean. However, the actuarial exams do have some (small) element of luck as well .I myself have not had a look at the assignment 6 for a while concentrating mostly on past papers and some wider reading.Of course, those who have had a look at the assignment prior to the exams and those who work in reserving (esp Mack/bootstrapping) , will have an advantage over the others!
Just for those panicking about bootstrapping A quick sample of reserving collegues (qualified or otherwise) suggests that they would have struggled with the question. Reserve ranges are very "new" to mainstream reserving, probably only getting used in reports etc in the last year or two and the actual theory of it is not well understood. They are used in the reserving risk elements of a capital model, so perhaps those who work in capital should do better.
in any case 15 marks for mack and bootstrap is quite a lot whether you have worked in capital modelling or not (and I am using mack). to be honest I did suspect that something like this would come up. GN50 requires actuaries to consider uncertainty which is quite new. Btw a range is not a requirement, but to consider uncertainty is. I suspect assignment 6 is from a previous past paper. So the examiner would be copying a previous past paper and not acted notes- they have no idea what is included in the acted material.
