Survey-weighted U5M risk stratification pipeline for Bangladesh DHS
Rattachement africain : bd. Niveau de preuve : code pays fourni par la source.
Le résumé fourni par la source
Final submission-aligned public release of the survey-weighted and temporally validated under-five mortality (U5M) risk stratification pipeline for Bangladesh Demographic and Health Survey (DHS) child-records data. This archive corresponds to the manuscript and supplementary material using the BDHS 2014 train, BDHS 2017–18 validation, and BDHS 2022 locked-test design. The release includes the main pipeline, export and sensitivity scripts, LaTeX manuscript sources, supplementary material, and tabular outputs supporting reported results. The workflow integrates DHS sampling weights, cluster-aware validation, post-hoc calibration, bootstrap uncertainty estimation, decision-curve analysis, explainability, and CRSS-based secondary policy stratification. BDHS microdata are not redistributed in this archive. Researchers must obtain the underlying DHS child-recode files directly from the DHS Program under their data-access terms. GitHub repository: https://github.com/subtlemyst/bdhs-u5m-survey-weighted-ml
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.
Le contrôle bibliographique ouvert
Les institutions déclarées
Une affiliation ne permet pas de déduire la nationalité d’un auteur.