The first step, for a reference year, is to calculate age-specific mortality, namely the ratio π*i of the number of people who died at a given age i to the number of people aged i in the population during the year in question. This provides an estimate of the probability of dying at each age i* (for a reference year).
We assume that age-specific mortality rates remain constant throughout life. Calculating life expectancy E at birth is then straightforward: it is the mean lifespan based on all these proportions. In other words:
E = 1 × π1 + 2 × π2 + 3 × π3 +…
Life expectancy at birth therefore represents the mean lifespan—that is, the mean age at death—of a hypothetical cohort that would experience, at each age, the mortality conditions of the year in question. We can also say that E is the number of years a person would, in principle, be expected to live from birth if mortality conditions remained fixed at their reference-year levels.
