ABSTRACT
This chapter illustrates a novel application of a recently developed structured additive regression model to analyze census and demographic data in SSA. The approach is data driven and has rarely been applied in the Saharan Africa (SSA) region using census data. The central question of the study was to identify risk factors associated with child mortality, mainly going beyond individual factors by including other unobserved factors, such as the geographic location where the child lives. These factors were assumed to vary in space and were best captured by assuming spatially varying processes. Our modeling approach can be considered as an extension of the Generalized Linear Models (GLM) and can be classified as a spatial GLM. These models have a complex structure, which is easily exploited using the Bayesian approach. Our analysis of child mortality shows evident continuity in African demography, with persistent cross-country heterogeneity, which is a salient feature in the global demographic profile.
