Functional Data Analysis Approach in Population Studies: An Application to the Gender Gap in Life Expectancy

Alessandro Feraldi , Sapienza University of Rome
Virginia Zarulli, University of Padova, Department of Statistics
Stefano Mazzuco , University of Padova
Cristina Giudici, Sapienza University of Rome

On average, women live longer than men and absolute difference between male and female mortality risk reaches its maximum at old ages. Although numerous studies have decomposed the sex gap in life expectancy according to age and causes of death, they did not study the main contributing components of the distributions of the age-specific contributions from a functional perspective, which is a more informative approach. To fill this gap, we study absolute and relative contributions of age and causes of death to the gender gap in life expectancy in 20 European and non-European countries between 1959 and 2015, using the Functional Data Analysis (FDA) and the Functional Principal Component Analysis (FPCA). Data were retrieved from the WHO Mortality Database and from the Human Mortality Database (HMD). Findings confirmed that lung-cancer, other neoplasms, diseases of circulatory system and external causes of death made the largest contributions to the GGLE in all the countries. FPCA allows to identify two main components that explain most of the variability and which capture the extent of the cause-specific gender differences and the age patterns, respectively. Over time, an increase in the most relevant contributions is observed, especially around the modal age and a shift of the contributions towards older age. First the study allows to identify country-specific patterns in the context of the epidemiological transition, which suggest the opportunity for further investigation. Additionally, it illustrates the demographic application of FDA, which could prove useful to deepen our understanding and knowledge of complex demographic phenomena.

See extended abstract

 Presented in Session P1. Postercafe