Hi Can we still use the term 'best estimate' on a point estimate if it was derived using partially credible information? Thanks.
Dear Jensen, Certainly we can. We would define "best estimate" as the actuary's view of the mean, or expected value of the outcome. In practice however, different people will have a different interpretation of what "best estimate" means. (In Subject CA1 for example, it is defined as the median outcome.) Therefore, we should be careful to explain what we mean when we use this term. With regards to you data question, in practice the data we use to derive our estimates is always flawed to some extent. The key point these days is how we communicate our best estimate. If the data is not fully credible, we would need to explain this (and the effect we think this could have) when we communicate our results. Indeed, Section 4.2 of GN12 explicitly tells us to do this. I would advise you to read GN12, and Section 4 of Chapter 16. Kind regards, Katherine.
Hi Katherine Is version 3.0 is the latest one? The profession's website search results turns out to be quite messy when you search for GN12.
Hi Jensen, I'm sorry for the slow reply, I'm afraid this post dropped down the list and I missed it. GN12 has been adopted by The Board for Actuarial Standards. Therefore the current version can be found on the BAS website. Here's the link: http://www.frc.org.uk/bas/actuarial/index.cfm Kind regards, Katherine.
