Comparing the Mortality Regimes in 39 Populations

The scope of this paper is to analyse mortality regimes and classify 39 populations using data from the Human Mortality Database. After utilising the life tables of these populations, several parameters will be studied: life expectancy at birth, Gini Coefficient, average life years lost because of deaths (e†), interquartile range (IQR), age separating early and late deaths, modal age at death and length of the old age heap. Afterwards, PCA analysis will produce uncorrelated components, which will be used in the subsequent cluster analysis to identify the homogeneous groups of populations, i.e., their segmentation. The study’s results indicate the existence of two mortality patterns in both genders. In males, these two major clusters are divided into three sub-clusters showing different transitional levels: one of more advanced, one of moderate and one of less advanced mortality transition. In females, four sub-clusters are formed, with several dissimilarities among them. The first two subclusters are of advanced mortality transition, and the second two are of less advanced. Details of this classification can be seen in the text. The segmentation of the populations differs in the two genders, signifying the differential patterns in their mortality regimes.