IDENTIFYING GLOBAL UNDER-FIVE MORTALITY HOTSPOTS BASED ON SPATIAL OUTLIER DETECTION AND ROBUST REGRESSION ESTIMATORS: POLICY INSIGHTS
DOI:
https://doi.org/10.33003/fjs-2026-1008-4885Keywords:
Outlier Detection, Robust Regression, Mortality Rate, EstimatorsAbstract
Under-five mortality remains a critical public health problem driven by multiple socioeconomic, health, and environmental factors. The severity of this problem varies across regions, depending on the underlying contextual characteristics. Understanding the spatial distribution and temporal pattern of under-five mortality can provide actionable insights for targeted interventions in vulnerable areas and help identify key sources of variation in mortality. In this study, we apply multivariate outlier detection and classification methods to identify countries at high risk of under-five mortality, using the robust multivariate CovMCD and cellMCD techniques. We exploit the concepts of regular, cellwise, and rowwise outliers to classify the risk levels of different countries. The analysis uses 2023 global under-five mortality data for 199 countries obtained from the United Nations Inter-agency Group for Child Mortality Estimation and evaluates robust regression methods for modelling global under-five mortality (U5MR). Our results show that spatial maps derived from CovMCD and cellMCD identify Africa as the most vulnerable region in terms of under-five mortality risk. Furthermore, we apply four regression estimators: ordinary least squares (OLS), least absolute deviations (LAD), least trimmed squares (LTS), and MM to compare their performance in the presence of cellwise and rowwise contamination. Both the real-data analysis and simulation study indicate that the MM estimator outperforms its competitors in terms of average coefficient RMSE and RMSPE across different outlier scenarios. Overall, our findings demonstrate that conventional OLS models can be misleading in the presence of multivariate outliers and reinforce the value of robust methods for policy-relevant child mortality modelling.
References
Cai, J., Hu, W., Yang, Y., Yan, H., & Chen, F. (2024). Outlier detection in spatial error models using modified thresholding-based iterative procedure for outlier detection approach. BMC Medical Research Methodology, 24(1), Article 89. doi.org
Chao, F., You, D., Pedersen, J., Hug, L., & Alkema, L. (2018). National and regional under-5 mortality rate by economic status for low-income and middle-income countries: A systematic assessment. The Lancet Global Health, 6(5), e535–e547. doi.org
Currit, N. (2002). Inductive regression: Overcoming OLS limitations with the general regression neural network. Computers, Environment and Urban Systems, 26(4), 335–353. doi.org
Dodge, Y. (1997). LAD regression for detecting outliers in response and explanatory variables. Journal of Multivariate Analysis, 61(1), 144–158. doi.org
Ezeh, O. K., Ogbo, F. A., Odumegwu, A. O., Oforkansi, G. H., Abada, U. D., Goson, P. C., Ishaya, T., & Agho, K. E. (2021). Under-5 mortality and its associated factors in northern Nigeria: Evidence from 22,455 singleton live births (2013–2018). International Journal of Environmental Research and Public Health, 18(18), Article 9899. doi.org
Filzmoser, P., & Gregorich, M. (2020). Multivariate outlier detection in applied data analysis: Global, local, compositional and cellwise outliers. Mathematical Geosciences, 52(8), 1049–1066. doi.org
Filzmoser, P., Höppner, S., Ortner, I., Serneels, S., & Verdonck, T. (2020). Cellwise robust M regression. Computational Statistics & Data Analysis, 147, Article 106944. doi.org
Fitrianto, A., & Xin, S. H. (2022). Comparisons between robust regression approaches in the presence of outliers and high leverage points. BAREKENG: Jurnal Ilmu Matematika dan Terapan, 16(1), 243–252. doi.org
Genç, M., & Lukman, A. (2025). Weighted LAD-Liu-LASSO for robust estimation and sparsity. Computational Statistics, 1–30. Advance online publication. doi.org
Gharehgozli, O. (2021). An empirical comparison between a regression framework and the synthetic control method. The Quarterly Review of Economics and Finance, 81, 70–81. doi.org
Giloni, A., Simonoff, J. S., & Sengupta, B. (2006). Robust weighted LAD regression. Computational Statistics & Data Analysis, 50(11), 3124–3140. doi.org
Horrace, W. C., & Oaxaca, R. L. (2006). Results on the bias and inconsistency of ordinary least squares for the linear probability model. Economics Letters, 90(3), 321–327. doi.org
Huang, D., Cabral, R., & De la Torre, F. (2015). Robust regression. IEEE Transactions on Pattern Analysis and Machine Intelligence, 38(2), 363–375. doi.org
Hu, Y., Fang, X., Zeng, W., Kutterer, H., & Li, D. (2024). Statistical robust estimation of spatial symmetric transformations based on Mahalanobis distance. IEEE Transactions on Geoscience and Remote Sensing, 62, 1–10. doi.org
Isiko, I., Jacob, E. C., Mwesigwa, A., Olot, H., Okoro, L. N., Asingwire, J. M., Izunwanne, M. J. P., Ikwara, E. A., Ajudua, S. C., Gmanyami, J. M., & Alonge, O. (2025). Socioeconomic, demographic and environmental factors associated with under-five mortality among children in Kenya: Analysis of the 2022 Kenya Demographic and Health Survey. BMC Pediatrics, 25(1), Article 509. doi.org
Ismail, M. I., & Rasheed, H. A. (2021). Robust regression methods/A comparison study. Turkish Journal of Computer and Mathematics Education, 12(14), 2939–2949. doi.org
Jiao, X., Pretis, F., & Schwarz, M. (2024). Testing for coefficient distortion due to outliers with an application to the economic impacts of climate change. Journal of Econometrics, 239(1), Article 105547. doi.org
Kaombe, T. M., & Manda, S. O. (2023). A novel outlier statistic in multivariate survival models and its application to identify unusual under-five mortality sub-districts in Malawi. Journal of Applied Statistics, 50(8), 1836–1852. doi.org
Kefale, B., Jancey, J., Gebremedhin, A. T., Belay, D. G., Pereira, G., & Tessema, G. A. (2026). Under-five mortality estimation methods: A methodological systematic review. Annals of Epidemiology, 113, 71–77. doi.org
Khan, D. M., Ali, M., Ahmad, Z., Manzoor, S., & Hussain, S. (2021). A new efficient redescending M-estimator for robust fitting of linear regression models in the presence of outliers. Mathematical Problems in Engineering, 2021, Article 3090537. doi.org
Khotimah, K., Sadik, K., & Rizki, A. (2020). Study of robust regression modeling using MM-estimator and least median squares. Journal of Physics: Conference Series, 1497(1), Article 012014. doi.org
Kiviet, J. F., & Phillips, G. D. (1996). The bias of the ordinary least squares estimator in simultaneous equation models. Economics Letters, 53(2), 161–167. doi.org
Lakshmi, R., & Sajesh, T. (2023). Empirical study on robust regression estimators and their performance. Reliability: Theory & Applications, 18(2), 466–478. doi.org
Lei, L., & Wooldridge, J. (2022). What estimators are unbiased for linear models? (ArXiv: 2212.14185). arXiv Preprint. arxiv.org
Musa, M., Asiribo, O., Dikko, H., Usman, M., & Sani, S. (2020). Modelling the determinants of under-five child mortality rates using Cox proportional hazards regression model. FUDMA Journal of Sciences, 4(4), 401–408. doi.org
Mwanga, M. K., Mirau, S. S., Tchuenche, J. M., & Mbalawata, I. S. (2025). Bayesian prediction of under-five mortality rates for Tanzania. Franklin Open, 10, Article 100221. doi.org
Pitselis, G. (2002). Application of GM and MM estimators to regression credibility. In Proceedings of the 2nd Conference in Actuarial Science and Finance (pp. 85–104). University of the Aegean. aegean.gr
Portnoy, S. (2022). Linearity of unbiased linear model estimators. The American Statistician, 76(4), 372–375. doi.org
Prahutama, A., & Rusgiyono, A. (2021). Robust regression with MM-estimator for modelling the number maternal mortality of pregnancy in Central Java, Indonesia. Journal of Physics: Conference Series, 1943(1), Article 012148. IOP Publishing. doi.org
Raymaekers, J., & Rousseeuw, P. J. (2026). Challenges of cellwise outliers. Econometrics and Statistics, 38, 6–25. https://doi.org/10.1016/j.ecosta.2024.02.002
Raza, A., Talib, M., Noor-ul Amin, M., Gunaime, N., Boukhris, I., & Nabi, M. (2024). Enhancing performance in the presence of outliers with redescending M-estimators. Scientific Reports, 14(1), Article 13529. doi.org
Rousseeuw, P. J., & Bossche, W. V. D. (2018). Detecting deviating data cells. Technometrics, 60(2), 135–145. doi.org
Rousseeuw, P. J., & Van Driessen, K. (2006). Computing LTS regression for large data sets. Data Mining and Knowledge Discovery, 12(1), 29–45. doi.org
Shatz, I. (2024). Assumption-checking rather than (just) testing: The importance of visualization and effect size in statistical diagnostics. Behavior Research Methods, 56(2), 826–845. doi.org
Shitu, S., Weldesamuel, B., Guesh, M., Abrha, S., Woldegerima, H., & Assefa, N. (2025). Spatial patterns and determinants of under-five mortality in Sub-Saharan Africa: A multi-country analysis. BMC Public Health, 25(1), Article 341. doi.org
Silva, R. (2012). Child mortality estimation: Consistency of under-five mortality rate estimates using full birth histories and summary birth histories. PLoS Medicine, 9(8), Article e1001296. doi.org
Stocker, T. C. (2008). On the asymptotic properties of the OLS estimator in regression models with fractionally integrated regressors and errors [Doctoral dissertation, University of Konstanz]. KOPS Institutional Repository. uni-konstanz.de
Su, P., Tarr, G., Müller, S., & Wang, S. (2024). CR-lasso: Robust cellwise regularized sparse regression. Computational Statistics & Data Analysis, 197, Article 107971. doi.org
Tagoe, E. T., Agbadi, P., Nakua, E. K., Duodu, P. A., Nutor, J. J., & Aheto, J. M. K. (2020). A predictive model and socioeconomic and demographic determinants of under-five mortality in Sierra Leone. Heliyon, 6(3), Article e03508. https://doi.org/10.1016/j.heliyon.2020.e03508
Ugrinowitsch, C., Fellingham, G. W., & Ricard, M. D. (2004). Limitations of ordinary least squares models in analyzing repeated measures data. Medicine & Science in Sports & Exercise, 36(12), 2144–2148. doi.org
United Nations Inter-agency Group for Child Mortality Estimation. (2024). Levels & trends in child mortality: Report 2024. United Nations Children's Fund. childmortality.org
Van Malderen, C., Amouzou, A., Barros, A. J., Masquelier, B., Van Oyen, H., & Speybroeck, N. (2019). Socioeconomic factors contributing to under-five mortality in Sub-Saharan Africa: A decomposition analysis. BMC Public Health, 19(1), Article 760. doi.org
Wang, L. (2003). Determinants of child mortality in LDCs: Empirical findings from Demographic and Health Surveys. Health Policy, 65(3), 277–299. doi.org
Wilcox, R. R. (2012). Introduction to robust estimation and hypothesis testing (3rd ed.). Academic Press. doi.org
World Health Organization. (2026). Child mortality (under 5 years). WHO Fact Sheets. who.int
Yildirim, V., & Mert Kantar, Y. (2020). Robust estimation approach for spatial error model. Journal of Statistical Computation and Simulation, 90(9), 1618–1638. doi.org
Zdaniuk, B. (2024). Ordinary least-squares (OLS) model. In A. C. Michalos (Ed.), Encyclopedia of quality of life and well-being research (pp. 4867–4869). Springer. doi.org
Downloads
Published
Issue
Section
Categories
License
Copyright (c) 2026 Mohammed Bappah Mohammed, Ishaq Abdullahi Baba, Rabiu Mohammed Madaki
This work is licensed under a Creative Commons Attribution 4.0 International License.
