Decomposing the Changing Sex-Gap in Life Expectancy: An Extension Method; and Mortality Forecasting with Modified Lee-Carter Method: Adjusting for Smoothing and Lifespan Disparity

In developed countries, males have higher mortality than females at almost all ages. Based on the recorded data, numerous studies observed that in many western countries the sex-gap in life expectancy at birth widens from the 1950s to the 1970s, followed by the sex-gap narrowing since the 1980s. Explanations of this pattern come from the epidemiologic transition, cardiovascular revolution, biological differences, and behavioural changes. Using decomposition methods for the sex difference in life expectancy, many studies have identified the ages and causes of death contribution to this gap. However, research into the reasons why the sex-gap changes over time is relatively sparse. To achieve a deeper understanding of the changing sex-gap in life expectancy at birth, we extend a decomposition method to include information on: the improvements in mortality, variability of deaths and survival advantage.

Qi Cui is a research assistant at the School of Demography, ANU. He finished his Master in Social Research at ANU in 2017. His current research focuses on mortality.

Extrapolative methods like the Lee-Carter method and its later variants are widely accepted for forecasting mortality in industrial countries due to simplicity. We propose to apply the Lee-Carter method on smoothed mortality rates obtained by LASSO type regularisation
and hence to partially adjust the time component of the Lee-Carter model to match the observed lifespan disparity. Smoothing by lasso produces fewer error during the fitting period compared to other spline based smoothing techniques. As a more informative indicator of
longevity, partially matching with observed lifespan disparity made the time component more reflective of the mortality improvement. The suggested modification generates fewer forecast errors for mortality rates for lower mortality countries and more accurate life expectancy forecasting for comparatively high mortality regime.

Ahbab Mohammed Fazle Rabbi is a PhD student at the Department of Statistical Sciences, University of Padua, Italy. Rabbi had his Masters in Statistics, Biostatistics and Informatics from the Department of Statistics, University of Dhaka, Bangladesh. His current research is concentrated on mortality forecasting.

Image gallery

Date & time

Fri 23 Mar 2018, 3:00pm to 4:00pm


Jean Martin Room, Beryl Rawson Bldg 13, Ellery Crescent, ANU


Qi Cui and Fazle Rabbi


Susan Cowan
6125 4273


Updated:  20 March 2018/Responsible Officer:  Head of School/Page Contact:  CASS Marketing & Communications