Forecasting age distribution of life-table death counts via <i>α</i> -transformation (Q6975101)

From a demographic perspective, mortality forecasts are an essential part of forward planning for national health and aged care systems and are an important component of population forecasts. In demography, three functions related to mortality are widely studied: central mortality rates, survival probabilities, and life-table death counts. \N\NThe authors consider life-table death counts and introduce a compositional power transformation, known as an \(\alpha\)-transformation, to model and forecast a time series of life-table death counts, possibly with zero counts observed at older ages. As a generalisation of the isometric log-ratio transformation, the \(\alpha\)-transformation relies on the tuning parameter \(\alpha\) which can be determined in a data-driven manner. Using the Australian age-specific period life-table death counts from 1921 to 2020, the authors show that \(\alpha\)-transformation can produce more accurate short-term point and interval forecasts than the log-ratio transformation.