Is ibnr_ a misnomer when it refers to an unpaid claim reserve?

[genedan]

When .ibnr_ is fitted from paid losses, it would really mean the unpaid claims reserve. I'm wondering if simply calling it .reserve_ would be better choice, as it is input neutral (or if this issue has been discussed/settled before)?

[henrydingliu]

yeah ken hates it. but i don't know what's a better term

at the end of the day, i think actuaries deal with naming difference all the time that it's not a big deal. everyone still uses incurred to refer to reported.

[genedan]

+1 vote for me hating it. For now, a non-invasive change would be to clarify what it is in the docs. For example, this snippet from the BF method could be changed:

to saying it's the reserve of the respective column, i.e., ultimate X - X where X is the column. The docstring for ultimate can be changed too, for example ultimate sal/sub would be something different from just losses.