Forecasting mortality in the event of a structural change

Edviges Coelho, Statistics Portugal
Luis Nunes, Universidade Nova de Lisboa

In recent decades, life expectancy in developed countries has risen to historically unprecedented levels driven by an unforeseen decline in mortality rates. The prospects of further reductions are of fundamental importance in various areas. In this context, this paper proposes a new approach to forecast future mortality and life expectancy in the event of a structural change. We show how recent advances in statistical testing for structural changes can be used to arrive at a properly specified forecasting model. Specifically, the results of tests for a change in the trend of the general level of mortality and for the presence of a unit-root are used to identify the appropriate model to be estimated and used to carry out the projections. The proposed methodology is applied to post-1950 time series mortality data for 18 developed countries. Structural changes in the rate of decline in overall mortality are found for almost every country considered and especially in the male populations. These are associated with a more accentuated decline in mortality in recent years. We also illustrate how accounting for such a change can lead to a major impact in mortality and life expectancy forecasts over the next decades.