A logistic two-population mortality projection model for modelling mortality at advanced ages for both sexes

In the current literature, numerous mortality projection models have been proposed and tested, but in general they have been designed for and applied to mainly ages below 90. As medical advances are being shifted to older ages over time and there is a rapid growth in the number of centenarians, there is a need to expand the modelling to older ages. We propose a logistic two-population mortality projection model for the death rates at ages 80 to 100+ for both sexes. We apply this model and its extensions to high quality old-age mortality data of Belgium, Sweden, Switzerland, and the UK and produce decent model performance in both mortality fitting and forecasting. The model structure also provides a reasonable way to close off the life table, which is supported by both theoretical arguments and empirical evidence.