Berq Sherm Paid method
Hi,
Small conceptual doubt.
If the underlying trend for the purposes of applying reported berq-sherman method is calculated using the paid severity triangle in case of changes to case reserve adequacy, then why can't we calculate the very same trend from the average case outstanding triangle for the purposes of applying the paid berq-sherman method in cases of changes to settlement rates ONLY? Wouldn't that make the workings so much easier than interpolation step involved.
Thanks
Keshav
Comments
It sounds like you're talking about restating (detrending) the paid loss triangle directly based on a trend derived from the average case outstanding, without the intermediate step of restating the paid counts. I'm not sure whether or not that would work. If you think you have an alternate method, you should make up and example and try it to see if you get an answer that's close to the answer you would get from the standard method.
One thing to think about however is that for the reported B-S method, the distortion is due to the case outstanding amount so that's what gets adjusted. For the paid B-S method, the distortion is due to the paid count triangle (assuming the final paid losses don't change - they are just paid earlier or later than before) so the B-S method adjusts the paid counts directly since that's the source of the problem. Using your method, you sidestep the issue with the paid counts and restate the paid losses directly. Like I said, I'm not sure if it would work - you can give it a try and find out. It's perfectly possible there are valid alternate methods.