Coherent Mortality Forecasting for Less Developed Countries

This paper proposes a multi-population approach to forecasting mortality rates for the less developed countries in a coherent way. The majority of these countries have witnessed faster mortality declines among the young and the working age population, whereas in developed countries, the mortality declined more heavily among the elders over the same period. We argue that the long-term mortality forecasts of a less developed country based solely on its historical mortality pattern might be implausible. As an alternative, we propose to incorporate the mortality data of a group of more developed countries as the benchmark to improve the forecasts for a less developed country. With the long-term coherence in mind, we propose to let a less developed country's age-specific mortality improvement rates to gradually converge to the benchmark values estimated from a group of more developed countries during the projection phase. Moreover, we employ a data-driven and threshold hitting approach to control the speed of this convergence. Empirical applications to China, Brazil, and Nigeria provide more reasonable long-term mortality forecasts than the existing forecasting practice