Modeling and forecasting sex differences in mortality: a sex-ratio approach

Marie Pier Bergeron-Boucher*, Vladimir Canudas-Romo, Marius Pascariu, Rune Lindahl-Jacobsen

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    13 Citations (Scopus)


    Female and male life expectancies have converged in most industrialized societies in recent decades. To achieve coherent forecasts between females and males, this convergence needs to be considered when forecasting sex-specific mortality. We introduce a model forecasting a matrix of the age-specific death rates of sex ratio, decomposed into two age profiles and time indices—before and after age 45—using principal component analysis. Our model allows visualization of both age structure and general level over time of sex differences in mortality for these two age groups. Based on a prior forecast for females, we successfully forecast male mortality convergence with female mortality. The usefulness of the developed model is illustrated by its comparison with other coherent and independent models in an out-of-sample forecast evaluation for 18 countries. The results show that the new proposal outperformed the other models for most countries.

    Original languageEnglish
    Article number20
    Issue number1
    Publication statusPublished - 1 Dec 2018


    Dive into the research topics of 'Modeling and forecasting sex differences in mortality: a sex-ratio approach'. Together they form a unique fingerprint.

    Cite this