Classification trees as a tool for decomposing inequalities in under -five mortality in Ghana

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
The study explores the socio-economic and demographic inequalities that exist in under- ve mortality using Ghana DHS 2008 data.These inequalities were inves-tigated using Classi cation trees developed by Breiman et al (CART). The data was analyzed using the Rpart of the R software 3.01 At the end of the analysis, total number of children ever born by mothers, pre-ceding birth interval, region of residence, ethnicity, highest educational level and wealth index were identi ed to contribute positively to these inequalities in under- ve mortality. It was also observed that Sex of a child and whether a child was born by cesarean does not contribute to these inequalities in the under- ve mor-tality in the GDHS data.
A thesis submitted to the Department of Mathematics, Kwame Nkrumah University of Science and Technology in partial fufillment of the requirement for the degree of M.phil Mathematical Statistics.