Spring 2017, Q17
Hi,
Just a small question. It has been specifically told in the question that the law change reduced the severity from 2016 "AY" and the reported figures for 2016 already incorporate that. That means there's no adjustment required and assuming the development pattern still stays the same as before, we just need to apply FS technique #1. As in the "extra" info should have no impact on my calculations as shown by Sample Answer 2.
Isn't any other approach like in Sample answer 1 and 3, where any adjustment for severity HAS been done, is literally wrong? And why had they been given credit when anything other than that followed/assumed in Sample answer 2 has been called a common mistake in the examiners' comments?
And if it is wrong, then how are the answers from all 3 approaches surprisingly close?
Thanks
Comments
Sample answers 1 & 2 give the same result because the actual severities for AY 2014 and 2015 were irrelevant. As you noted, the development pattern did not change so whether or not you multiply the reported losses by 0.75 for AY 2014 & 2015, you will get the same LDFs and therefore the same final answer.
Strictly speaking, I would say sample answer 2 is not completely correct because it doesn't adjust for the law change for AY 2014 & 2015 but mathematically it didn't matter. (Every entry in the 2014 row was adjusted by the same factor. Same for 2015.)
I would ignore sample answer 3. It calculates incremental triangles but that's an unnecessary complication. This was intended as a fairly straightforward application of frequency-severity method 1 (aside from the minor complication of the law change.)