Sa3 September 2006

[swarren]

Any thoughts on this paper??

I felt very rushed for time, especially as I felt you could have easily spent too much time on the second question.

Yet there were some parts where I could hardly think of anything to write well not as much as the marks implied.

 

[mark]

sa3 sep 2006

I thought the last two questions - data for risk premium and then other data for underwriting was a bit off - some issues you could place under either I thought. Thats not a big complaint I guess.

Describe how you would use the benchmark data to est the reserve was a bit strange wording I thought (think this was 1iii). I guess for that many marks (was it 14?) you'd want to do more than "describe", I thought you'd probably want to actually do it. I on-leveled the latest bm ult loss ratios and got an average of 57% or something, and concluded that with only 6m experience you'd probably just reserve it at 60% or something. In any case I ended up both describing various ways of doing it and then doing it the most obvious way given the data they gave. It will be very annoying if they only meant us to do one of those things and just worded the q badly.

Found it very hard to think of things to say in last part of q1 which was "How you would go about reserving at year-end 2006" if I remember correctly. Short of pulling stats about how developed the 2005 and 2006 cohorts would be then, and suggest BF might not be a bad idea I found it hard to come up with stuff for 8 marks (was it 8?).

What did other people do?

 

[Niel]

I agree with you Swarren, it was a bit rushed. By the middle of Q1 I was 10 mins behind the clock and playing catch-up from then on. That is not good for generating ideas.

Mark, vague questions are examinators' best weapon. Even going through some past papers I would read a question and have no idea what they are looking for. Then looking at the answer I would think "Oh, was that all". Part of my preperation is actually to analyse questions and see how they got to their answers.

I had some trouble generating enough ideas for some points but at other had to leave out some ideas. For example for "Suggest risks and ways of mitigating them" I could probably write 40 marks.

I must admit, I can't remember exactly what I did at each question, but B-F did sound like a good idea for the reserving Q given all ultimate loss ratio benchmark. I also mentioned the 6 month, 18 month ... development years might be a problem.

Now for the long wait...