Abstract
Malaria and anaemia which jointly account for high proportion of morbidity and mortality among young children in developing countries have been individually studied using binary regression model. We adopt geoadditive latent variable model for binary/ordinal indicators to analyze the influence of variables of different types on the morbidity among young children in Nigeria. Latent variable models allow for the analysis of multidimensional response variables that reveal the indicator's underlying relationship that are caused by the latent variables. We extend the structural model to a semi-parametric geoadditive model in order to quantify the joint spatial structure of morbidity from malaria and anaemia. Findings revealed substantial geographical variations and the generated maps can guide policy makers and donors on how to prudently utilize the scarce resources for designing more cost-effective interventions. (C) 2016 Elsevier Ltd. All rights reserved.
Dokumententyp: | Zeitschriftenartikel |
---|---|
Fakultät: | Mathematik, Informatik und Statistik > Statistik |
Themengebiete: | 500 Naturwissenschaften und Mathematik > 510 Mathematik |
ISSN: | 1877-5845 |
Sprache: | Englisch |
Dokumenten ID: | 47317 |
Datum der Veröffentlichung auf Open Access LMU: | 27. Apr. 2018, 08:12 |
Letzte Änderungen: | 04. Nov. 2020, 13:24 |