Reporting delays

(Jan 19, 2021)

When performing a mortality analysis, it is my practice to disregard the most recent six months or so of experience data.  The reason is delays in the reporting and recording of  deaths, i.e. occurred-but-not-reported (OBNR) to use the terminology of Lawless (1994).  We use the term OBNR, rather than the more familiar term IBNR (incurred-but-not-reported); IBNR is associated with "cost-orientated" delay distributions of insurance claims (Jewell, 1989), whereas we are focused on just the delay itself.

But what evidence is there to support my six-month rule of thumb?  As part of ongoing research in Richards (2021), I received separate extracts of annuitant mortality experience in June and September 2020

Read more

Tags: OBNR, IBNR, right truncation, interval censoring

Find by key-word


Find by date


Find by tag (show all )