Analyzing Cause-Specific Mortality Trends using Compositional Functional Data Analysis

TL;DR

This paper introduces a new compositional functional data analysis framework to study cause-specific mortality trends across countries, revealing key variations and gender differences in mortality patterns over time.

Contribution

The paper develops a novel functional PCA-based method for compositional mortality data, enabling detailed analysis and clustering of cause-specific mortality trends.

Findings

Main modes of variation in mortality rates identified

Clustering reveals distinct country groups based on mortality trends

Gender differences in cause-specific mortality evolution highlighted

Abstract

We study the dynamics of cause--specific mortality rates among countries by considering them as compositions of functions. We develop a novel framework for such data structure, with particular attention to functional PCA. The application of this method to a subset of the WHO mortality database reveals the main modes of variation of cause--specific rates over years for men and women and enables us to perform clustering in the projected subspace. The results give many insights of the ongoing trends, only partially explained by past literature, that the considered countries are undergoing. We are also able to show the different evolution of cause of death undergone by men and women: for example, we can see that while lung cancer incidence is stabilizing for men, it is still increasing for women.