Volume 30 - Article 35 | Pages 1011–1034 Author has provided data and code for replicating results

Regional probabilistic fertility forecasting by modeling between-country correlations

By Bailey Fosdick, Adrian E. Raftery

Print this page  Facebook  Twitter


Date received:01 Dec 2012
Date published:02 Apr 2014
Word count:5730
Keywords:aggregation, correlation, forecasting, probabilistic projection, pseudolikelihood, total fertility rate (TFR)
Additional files:readme.30-35 (text file, 1 kB)
 demographic-research.30-35 (zip file, 124 kB)


Background: The United Nations (UN) Population Division constructs probabilistic projections for the total fertility rate (TFR) using the Bayesian hierarchical model of Alkema et al. (2011), which produces predictive distributions of the TFR for individual countries. The UN is interested in publishing probabilistic projections for aggregates of countries, such as regions and trading blocs. This requires joint probabilistic projections of future countryspecific TFRs, taking account of the correlations between them.

Objective: We propose an extension of the Bayesian hierarchical model that allows for probabilistic projection of aggregate TFR for any set of countries.

Methods: We model the correlation between country forecast errors as a linear function of time invariant covariates, namely whether the countries are contiguous, whether they had a common colonizer after 1945, and whether they are in the same UN region. The resulting correlation model is incorporated into the Bayesian hierarchical model’s error distribution.

Results: We produce predictive distributions of TFR for 1990-2010 for each of the UN's primary regions. We find that the proportions of the observed values that fall within the prediction intervals from our method are closer to their nominal levels than those produced by the current model.

Conclusions: Our results suggest that a substantial proportion of the correlation between forecast errors for TFR in different countries is due to the countries' geographic proximity to one another, and that if this correlation is accounted for, the quality of probabilistic projections of TFR for regions and other aggregates is improved.

Author's Affiliation

Bailey Fosdick - Duke University, United States of America [Email]
Adrian E. Raftery - University of Washington, United States of America [Email]

Other articles by the same author/authors in Demographic Research

» Probabilistic forecasting of maximum human lifespan by 2100 using Bayesian population projections
Volume 44 - Article 52

» Probabilistic projection of subnational total fertility rates
Volume 38 - Article 60

» Bayesian projection of life expectancy accounting for the HIV/AIDS epidemic
Volume 37 - Article 48

» Joint probabilistic projection of female and male life expectancy
Volume 30 - Article 27

» Estimating trends in the total fertility rate with uncertainty using imperfect data: Examples from West Africa
Volume 26 - Article 15

Most recent similar articles in Demographic Research

» Probabilistic projection of subnational total fertility rates
Volume 38 - Article 60    | Keywords: correlation, total fertility rate (TFR)

» Leveraging deep neural networks to estimate age-specific mortality from life expectancy at birth
Volume 47 - Article 8    | Keywords: forecasting

» The contribution of assisted reproductive technology to fertility rates and parity transition: An analysis of Australian data
Volume 45 - Article 35    | Keywords: total fertility rate (TFR)

» How do populations aggregate?
Volume 44 - Article 15    | Keywords: aggregation

» Is a positive association between female employment and fertility still spurious in developed countries?
Volume 41 - Article 45    | Keywords: total fertility rate (TFR)