The overarching goal of the BayesPop working group is to develop a statistical framework for probabilistic population projections for all countries. On July 11, 2014 (World Population Day), the United Nations issued official probabilistic population projections for all countries for the first time, using our new statistical methods. A previous experimental version had been issued by the UN in November 2012. Population projections have long been done in a deterministic way, and this is a major step towards putting them on a statistical and probabilistic basis.The specific aims of our further research are:
- Developing methods for probabilistic projection of correlated population quantities
- Developing methods for probabilistic projections of international migration
- Developing methods for age-specific probabilistic projections
- Probabilistic projections for countries with generalized HIV/AIDS epidemics
- Developing methods for reconstructing past populations with uncertainty from fragmentary data
- Producing publicly available software that implements the new methods
If you have a question about the methodology or software, please send a message or subscribe to our mailing list.
- Our work was featured in a documentary about demography on German national TV (3sat/ZDF). The segment about our work starts at 26:30.
- New versions of R packages bayesTFR, bayesLife and bayesPop are available on CRAN. They work now with wpp2015.
- wppExplorer has been updated to work with WPP 2015. Browse data online. No R necessary!
- Following the release of WPP 2015, the R package wpp2015 is now available on CRAN.
- WPP 2015 has been released on July 29th 2015.
- The work attracted some media coverage. For example, here are reports on it in The Guardian, and the Scientific American. Here and here are some comments on it in the New York Times. Here is Science magazine's own report on the research article. The article was ranked in the top 0.1% of over 30,000 articles ever published in Science by the journal's article level metrics.