The Impact of Multiple Structural Changes on Mortality Predictions

Most mortality models proposed in recent literature rely on the standard ARIMA-framework (in particular: a random walk with drift) to project mortality rates. As a result the projections are highly sensitive to the calibration period. We apply a modelling strategy for the time-dependent parameters in a large collection of mortality models that allows for objective, statistical detection of one or more structural changes. By comparing projections based on di fferent calibration periods and diff erent time series specifi cations, we show that our proposed methodology leads to more robust mortality projections