Ahbab Mohammad Fazle Rabbi , University of Dhaka
Stefano Mazzuco, University of Padova
Heather Booth, Australian National University
Vladimir Canudas-Romo, Australian National University
Coherent mortality forecasting methods try to capture the influence of global improvement of health, communication, science on a specific population. The widely used coherent model is an hierarchical, extended form of the Lee-Carter method which assumes an invariant age component and a presumably linear time component to model the joint mortality data of relational populations. Besides forecast inaccuracy due to estimation procedure, choosing the appropriate reference population remains an arbitrary process. We propose to apply the Lee-Carter method on smoothed mortality rates obtained by LASSO type regularization and hence to partially adjust the time component according to observed lifespan disparity to get the common factor of the relational populations. Time variability is also taken into consideration during obtaining the common factor. The reference group for making coherent forecast for a particular population is chosen from the set of available populations on the basis of closest lifespan disparity over time. The proposed methodology generates less forecast errors during out-of-sample forecast period and more optimistic forecast of life expectancy for most of the low-mortality countries. Moreover, choosing the relational populations on the basis of historical pattern of lifespan disparity made the choice of reference populations more systematic and competent.
Presented in Session 31. Mortality Predictions