Application of Bayesian Spatiotemporal Logistic Regression to Child Mortality in Nigeria

This study investigates the persistent challenge of under-five mortality in Nigeria despite notable global progress in reducing child deaths. Conventional models often assume homogeneous relationships between risk factors and outcomes, neglecting spatial and temporal heterogeneity. To address this limitation, we develop a Bayesian Geographically and Temporally Weighted Logistic Regression (GTWLR) framework an extension of the Geographically and Temporally Weighted Regression (GTWR) model to binary outcomes. The Bayesian GTWLR model incorporates spatial and temporal weighting, prior knowledge, and full uncertainty quantification. Model estimation was implemented using the Integrated Nested Laplace Approximation (INLA) for computational efficiency and robust inference. Using data from the Nigeria Demographic and Health Surveys (NDHS), the model captures local variations in under-five mortality, identifies high-risk regions, and quantifies uncertainty through posterior credible intervals.This approach offers a rigorous statistical foundation for evidence-based policymaking, enabling geographically targeted interventions aligned with Sustainable Development Goal (SDG) 3.2.

Child mortality; Bayesian inference; Spatiotemporal modeling; GTWLR; INLA

1. Fotheringham, A. S., Brunsdon, C., & Charlton, M. (2002). Geographically Weighted Regression: The Analysis of Spatially Varying Relationships. Chichester: Wiley. [Google Scholar] [Crossref]

2. Huang, B., Wu, B., & Barry, M. (2010). Geographically and temporally weighted regression for modeling spatio-temporal variation in house prices. International Journal of Geographical Information Science, 24(3), 383–401. doi:10.1080/13658810802672469 [Google Scholar] [Crossref]

3. Wu, B., Li, R., & Huang, B. (2014). A geographically and temporally weighted autoregressive model with application to housing prices. International Journal of Geographical Information Science, 28(5),1186-1204 doi:10.1080/13658816.2013.878463 [Google Scholar] [Crossref]

4. NBS & UNICEF. (2021). Multiple Indicator Cluster Survey 2021. Abuja: National Bureau of Statistics and UNICEF. [Google Scholar] [Crossref]

5. Harianto, W. H. Nugroho, & E. Sumarminingsih. (2021). Geographically and Temporally Weighted Regression Model with Gaussian Kernel Weighted Function and Bisquare Kernel Weighted Function. ICSTEIR 2020. IOP Conference Series: Materials Science and Engineering, 1115(012063). [Google Scholar] [Crossref]

6. Ohyver, M., Purhadi, A., & Choiruddin, A. (2025). Parameter Estimation of Geographically and Temporally Weighted Elastic Net Ordinal Logistic Regression. Mathematics, 13(13), 1345. doi:10.3390/math13131345 [Google Scholar] [Crossref]

7. Wang, S., Ren, Z., & Liu, X. (2023). Spatiotemporal trends in neonatal, infant, and child mortality (1990–2019) based on Bayesian spatiotemporal modeling. Frontiers in Public Health, 11, 996694. doi:10.3389/fpubh.2023.996694. https://www.frontiersin.org/articles/10.3389/fpubh.2023.996694/full [Google Scholar] [Crossref]

8. Egbon, Osafu Augustine; Bogoni, Mariella Ananias; Babalola, Bayowa Teniola; Louzada, Francisco. (2022). Under age five children survival times in Nigeria, a Bayesian spatial modeling approach. BMC Public Health, 22, 2207 [Google Scholar] [Crossref]

9. Lu, C., Black, M., & Richter, L. (2016). Risk of Poor Development in Young Children in Low-Income and Middle Income Countries: An Estimation and Analysis at the Globe, Regional, and Country Level. The Lancet Global Health, 4, e916-e922. [Google Scholar] [Crossref]

10. WHO. (2020). Levels and Trends in Child Mortality: Report 2020. New York: United Nations Inter-agency Group for Child Mortality Estimation [Google Scholar] [Crossref]

11. Rue, H., Martino, S., & Chopin, N. (2009). Approximate Bayesian inference for latent Gaussian models using INLA. Journal of the Royal Statistical Society: Series B, 71(2), 319–392. [Google Scholar] [Crossref]

12. doi:10.1111/j.1467-9868.2008.00700.x [Google Scholar] [Crossref]

13. Martins, T. G., Simpson, D., Lindgren, F., & Rue, H. (2013). Bayesian computing with INLA: New features. Computational Statistics & Data Analysis, 67, 68–83. doi:10.1016/j.csda.2012.07.015 [Google Scholar] [Crossref]

14. Lindgren, F., Rue, H., & Lindstrom, J. (2011). An explicit link between Gaussian fields and Gaussian Markov random fields: The SPDE approach. Journal of the Royal Statistical Society: Series B, 73(4), 423–498. doi:10.1111/j.1467-9868.2011.00777.x [Google Scholar] [Crossref]

15. Adebowale, A. S., Yusuf, B. O., & Fagbamigbe, A. F. (2017). Determinants of under-five mortality in Nigeria: A multilevel analysis. Global Health Action, 10(1), 128–141. https://doi.org/10.1186/s12887-016-0742-3 [Google Scholar] [Crossref]

16. Abiodun, A., Ayodele, O., & Ishaq, I. (2023). Comparison of Cox and Extended Cox Models on Age at First Marriage among Nigerian Women. Nigerian Journal of Basic and Applied Sciences, 30(2), 126–133. https://doi.org/10.4314/njbas.v30i2.17 [Google Scholar] [Crossref]

17. AA Oyinloye, KP Ajewole, RO Olanrewaju, OJ Ayodele, (2025). Comparative Modeling of Time Series with Asymmetric Autoregressive Process. NIPES-Journal of Science and Technology, Research, 7 pp. 1873–1880 [Google Scholar] [Crossref]

• Interplay of Students’ Emotional Intelligence and Attitude toward Mathematics on Performance in Grade 10 Algebra
• Numerical Simulation of Fitzhugh-Nagumo Dynamics Using a Finite Difference-Based Method of Lines
• Fixed Point Theorem in Controlled Metric Spaces
• Usage of Moving Average to Heart Rate, Blood Pressure and Blood Sugar
• Exploring Algebraic Topology and Homotopy Theory: Methods, Empirical Data, and Numerical Examples