Skip to main content

Probabilistic Subnational Population Projections

Hana Sevcikova

Hana Sevcikova

Abstract:

Population projections provide predictions of future population sizes for an area. Historically, most existing population projections have been produced using deterministic or scenario-based approaches, and did not assess uncertainty about future population change. Starting in 2015, however, the United Nations has produced probabilistic population projections for all countries using a Bayesian approach. There is also considerable interest in subnational probabilistic population projections, for example at the state and county levels. These are needed by local governments for planning, by the private sector for strategic decision-making, and by researchers, particularly in the health and social science research on subnational variation and inequality. A direct application of the UN approach to the subnational context has not turned out to be fully satisfactory, because within-country correlations in fertility and mortality are generally larger than between-country ones, migration is not constrained in the same way, and there is a need to account for college and other special populations, particularly at the county level. We propose an extension to the national framework that deals with these challenges and gives accurate and well-calibrated forecasts and forecast intervals.

 

Hana Sevcikova is a senior research scientist at the CSSS. She works on developing methods for probabilistic population projection, national and subnational. She has developed various demographic R packages that the United Nations Population Division has been using to produce the World Population Prospects. She also works as a data scientist on land use modeling for the Puget Sound Regional Council.


Room
409