Here you can download trajectories of probabilistic projections of total fertility rate, male and female life expectancy at birth, and population. They were produced using the R packages bayesTFR, bayesLife and bayesPop. Results were adjusted so that country medians match the World Population Prospects 2017. The provided files contain 1000 trajectories for 18 time periods (until 2100) for 201 countries. For population projections, we are also including trajectories for selected aggregated regions. Countries and regions are identified using the UN numerical codes. Here is a code lookup table.
- Total fertility rates (23M) updated 01/19/2018
- Female life expectancy (25M)
- Male life expectancy (26M)
- Population (38M) updated 01/19/2018
- Population for aggregated regions (6.2M) updated 01/19/2018
If you publish work that uses these trajectories, please cite the following:
Raftery, A.E., Li. N., Ševčíková , H., Gerland, P. and Heilig, G.K. (2012). Bayesian probabilistic population projections for all countries. Proceedings of the National Academy of Sciences 109:13915-13921.
United Nations (2017). World Population Prospects: The 2017 Revision. Population Division, Dept. of Economic and Social Affairs, United Nations, New York, NY.
Note: These trajectories are not official products of the United Nations. They were produced by the University of Washington BayesPop research group using software produced by the group. There may be differences between these results and the World Population Prospects 2017.
Here are probabilistic trajectories of international net migration, as well as trajectories of total population that take probabilistic migration into account. Both are on the scale of thousands. The provided files contain 1000 trajectories for 18 time periods (from 2015 to 2100) for 201 countries. We also include trajectories for selected aggregated regions. The projected migration trajectories sum to zero over the 201 countries. Since about 30 small countries and territories with populations under 100,000 are excluded from the population projections, they are also excluded from the zero-sum requirement as well as from the aggregations. Countries and regions are identified using the UN numerical codes. Here is a code lookup table.
- International Net Migration (21M)
- Population (23M)
- Migration for aggregated regions (World excluded) (3.7M)
- Population for aggregated regions (4.4M)
Summary files (median, mean, 80% and 95% probability intervals), including aggregations:
If you publish work that uses these trajectories, please cite the following articles, which describe the methodology used to produce them:
Azose, J.J. and Raftery, A.E. (2015). Bayesian Probabilistic Projection of International Migration Rates. Demography 52:1627-1650.
Azose, J.J., Ševčíková, H., Raftery, A.E. (2016): Probabilistic population projections with migration uncertainty. Proceedings of the National Academy of Sciences 113:6460–6465.
This dataset on subnational total fertility rates has been compiled in 2012 by Patrick Gerland (with inputs from Julien Thillard, Thomas Spoorenberg and Danan Gu) based on national data sources for each country. The reliability of these data vary between countries, but for a majority of countries fertility estimates are based on birth registration data.
A reduced ascii version of the dataset can be used as input to bayesTFR to produce subnational TFR projections.
For projecting population using bayesPop, results of bayesTFR and bayesLife are needed which often can take a long time to run. Here you can download such results containing converged MCMCs of the TFR and life expectancy models, as well as predictions.