Modeling the Relationship Between Male and Female Mortality Patterns in Modern Populations

It is generally known that the age-specific probabilities of dying exhibit a typical pattern common in all human populations.
It is also extensively known that, in every population, males and females experience significantly different mortality levels.
However the gap between the mortality patterns of the two sexes significantly varies through ages exhibiting a pattern that
became typical since the middle of the 20th century. In this paper we present a parametric model for the description of this
pattern, which presents the log ratio between the male and female empirical probabilities of dying as a parametric function
of age. For evaluating the adequacy of the model proposed, we fit it to empirical data sets of a variety of populations and
time periods. This model can serve as a tool for comparisons of the sex-differentiated mortality between populations and through
time.