The United Nations publishes updated estimates and projections of the populations of all the world's countries, broken down by age and sex. These are widely used by international organizations, governments, the private sector and researchers, for example for climate modeling and for assessing progress towards the Sustainable Development Goals. The UN's previous projections were deterministic, and under the previous grant, we developed a fully probablistic projection methodology, which was adopted by the UN for its official projections for all countries in 2015. The new projections changed the understanding of the outlook for population, indicating that stabilization of world population is unlikely this century, largely because of the slowdown in fertility decline in high-fertility countries. We will improve our methodology by taking account of generalized HIV/AIDS epidemics. We will also develop new methods that account for the evolution of smoking, a major factor for mortality. Our methods are based on estimates of past population and vital rates, but these have measurement error, particularly in the more than half of countries without good vital registration system, We will extend our methods to take account of measurement error in estimating past fertility rates. We will also assess the possible effects on fertility of major policy initiatives focused on child survival, girl's education, family planning and the status of women in high fertility countries We will produce publicly available software for implementing our new methods. We will also conduct training courses and continue to maintain an email list for users of the methods.
Every two years, the United Nations produces projections of the populations of all countries, which are widely used by international organizations, governments, the private sector and researchers. Under the previous grant we developed new statistical methods for probabilistic population projections, which were adopted by the UN for their official projections for all countries in 2015. We will improve the methodology to take account of generalized HIV/AIDS epidemics, the impact of smoking, and measurement error in past population estimates, and to assess the possible effects of major policy initiatives on fertility rates in high-fertility countries.
| Hernández, Belinda; Raftery, Adrian E; Pennington, Stephen R et al. (2018) Bayesian Additive Regression Trees using Bayesian Model Averaging. Stat Comput 28:869-890 |
| Sharrow, David J; Godwin, Jessica; He, Yanjun et al. (2018) Probabilistic population projections for countries with generalized HIV/AIDS epidemics. Popul Stud (Camb) 72:1-15 |
| Scrucca, Luca; Raftery, Adrian E (2018) clustvarsel: A Package Implementing Variable Selection for Gaussian Model-Based Clustering in R. J Stat Softw 84: |
| Godwin, Jessica; Raftery, Adrian E (2017) Bayesian projection of life expectancy accounting for the HIV/AIDS epidemic. Demogr Res 37:1549-1610 |
| Raftery, Adrian E; Zimmer, Alec; Frierson, Dargan M W et al. (2017) Less Than 2 °C Warming by 2100 Unlikely. Nat Clim Chang 7:637-641 |
| Hung, Ling-Hong; Shi, Kaiyuan; Wu, Migao et al. (2017) fastBMA: scalable network inference and transitive reduction. Gigascience 6:1-10 |
| McCormick, Tyler H; Lee, Hedwig; Cesare, Nina et al. (2017) Using Twitter for Demographic and Social Science Research: Tools for Data Collection and Processing. Sociol Methods Res 46:390-421 |
| Onorante, Luca; Raftery, Adrian E (2016) Dynamic Model Averaging in Large Model Spaces Using Dynamic Occam's Window. Eur Econ Rev 81:2-14 |
| Šev?íková, Hana; Raftery, Adrian E (2016) bayesPop: Probabilistic Population Projections. J Stat Softw 75: |
| Wheldon, Mark C; Raftery, Adrian E; Clark, Samuel J et al. (2016) Bayesian population reconstruction of female populations for less developed and more developed countries. Popul Stud (Camb) 70:21-37 |
Showing the most recent 10 out of 56 publications
