2016/09/14 by David Sharrow, Sharrow, David J., Jessica Godwin +7
Decision Sciences · Economics, Econometrics and Finance · Social Sciences · #Applications (stat.AP) #FOS: Computer and information sciences #HIV/AIDS Impact and Responses #Insurance, Mortality, Demography, Risk Management #Methodology (stat.ME) #Other Statistics (stat.OT) #demographic modeling and climate adaptation
paper · pdf · doi:10.48550/arxiv.1609.04383
openalex publication_date 2016/09/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The United Nations (UN) issued official probabilistic population projections\nfor all countries to 2100 in July 2015. This was done by simulating future\nlevels of total fertility and life expectancy from Bayesian hierarchical\nmodels, and combining the results using a standard cohort-component projection\nmethod. The 40 countries with generalized HIV/AIDS epidemics were treated\ndifferently from others, in that the projections used the highly multistate\nSpectrum/EPP model, a complex 15-compartment model that was designed for\nshort-term projections of quantities relevant to policy for the epidemic. Here\nwe propose a simpler approach that is more compatible with the existing UN\nprobabilistic projection methodology for other countries. Changes in life\nexpectancy are projected probabilistically using a simple time series\nregression model on current life expectancy, HIV prevalence and ART coverage.\nThese are then converted to age- and sex-specific mortality rates using a new\nfamily of model life tables designed for countries with HIV/AIDS epidemics that\nreproduces the characteristic hump in middle adult mortality. These are then\ninput to the standard cohort-component method, as for other countries. The\nmethod performed well in an out-of-sample cross-validation experiment. It gives\nsimilar population projections to Spectrum/EPP in the short run, while being\nsimpler and avoiding multistate modeling.\n