Please use this identifier to cite or link to this item:
Title: Estimating the Under-Five mortality rate using a Bayesian hierarchical time series model
Authors: Alkema, L. 
Ann, W.L.
Issue Date: 28-Sep-2011
Source: Alkema, L., Ann, W.L. (2011-09-28). Estimating the Under-Five mortality rate using a Bayesian hierarchical time series model. PLoS ONE 6 (9) : -. ScholarBank@NUS Repository.
Abstract: Background: Millennium Development Goal 4 calls for a reduction in the under-five mortality rate by two-thirds between 1990 and 2015, which corresponds to an annual rate of decline of 4.4%. The United Nations Inter-Agency Group for Child Mortality Estimation estimates under-five mortality in every country to measure progress. For the majority of countries, the estimates within a country are based on the assumption of a piece-wise constant rate of decline. Methods and Findings: This paper proposes an alternative method to estimate under-five mortality, such that the underlying rate of change is allowed to vary smoothly over time using a time series model. Information about the average rate of decline and changes therein is exchanged between countries using a Bayesian hierarchical model. Cross-validation exercises suggest that the proposed model provides credible bounds for the under-five mortality rate that are reasonably well calibrated during the observation period. The alternative estimates suggest smoother trends in under-five mortality and give new insights into changes in the rate of decline within countries. Conclusions: The proposed model offers an alternative modeling approach for obtaining estimates of under-five mortality which removes the restriction of a piece-wise linear rate of decline and introduces hierarchy to exchange information between countries. The newly proposed estimates of the rate of decline in under-5 mortality and the uncertainty assessments would help to monitor progress towards Millennium Development Goal 4. © 2011 Alkema, Ann.
Source Title: PLoS ONE
ISSN: 19326203
DOI: 10.1371/journal.pone.0023954
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.


checked on Feb 27, 2018


checked on Feb 14, 2018

Page view(s)

checked on Mar 12, 2018

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.