Application of Non-Parametric, Semi-Parametric and Parametric Survival Methods on Infant Mortality Data of Bangladesh

Abstract

The statistical analysis of lifetime or survival data has become a topic of considerable interest of statisticians. The field has expanded
rapidly in recent years. In this study, we consider three different approaches to survival analysis. The Kaplan-Meier method as a nonparametric
method, Cox’s proportional hazards model as a semi-parametric method and accelerated failure time model as a parametric
method have been used in this study to analyze the infant mortality data of Bangladesh Demographic Health Survey (BDHS) 2004 to
assess the influence of several factors like sex of child, mother’s age, mother’s education, birth order and type of place of residence on
infant survival. It has been found that all these factors have significant influence on infant mortality.