Information Matrix

Actuaries tasked with analysing a portfolio's mortality experience face a gap between what has happened in the outside world and the data they actually work with.  The various difference levels are depicted in Figure 1.

Figure 1.  The actuarial data onion.

Everyone is familiar with the idea of a forecast. You have data on a phenomenon up to the current time, and want to forecast the phenomenon at some point in the future.  The most obvious example is the weather forecast, but forecasting is also required in pension and annuity work.  For example, when calculating reserves for pension payments, some kind of projection is required for future mortality improvements.

In a previous blog I outlined my six-month rule of thumb for discarding mortality experience affected by reporting delays. However, this can be awkward where there is a hard limit on how far back the experience data goes. For example, when a pension scheme switches administrator, or an insurer migrates business from one system to another, past mortality data is usually the first casualty.

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.