Figure 5
From: How much can we trust life tables? Sensitivity of mortality measures to right-censoring treatment
Log-mortality of Swedish males in 1970: raw data (circles), HMD data—raw data until age 95 and Kannisto-smoothed data from age 96 onwards (squares), and ΓGM fit (solid line). The three dashed horizontal lines (corresponding to censoring ages 94, 96 and 100) reflect the assumption that the hazard in the last open-age group in a life table is constant. If the observed death rate at the age at censoring overestimates the true force of mortality, remaining life expectancy will be overestimated/underestimated if the differences between areas A and B is negative/positive. If the observed death rate at the age at censoring underestimates the true force of mortality, remaining life expectancy will be overestimated.
