Reconstruction of life tables and age distributions for the population of China, by year, from 1953 to 1982

Little is known about the mortality trend of the population of mainland China since the early 1950's. China does not publish annual records of the number of deaths and population by age and sex, thus conventional ways of constructing a life table cannot be applied.; The current research uses the data from three census age distributions, annual age-specific fertility rates and the figures of the total population by year to estimate life tables and age distributions by year for the population of China from 1953 to 1982. First, a system of equations is developed for estimating the survival ratio functions of life tables for all years of the studied period. This system of equations, without any knowledge of the age pattern of mortality, fails to give a finite solution but rather yields an infinite number of solutions. This is because the number of unknowns in the system is larger than the number of equations. Therefore it is necessary to introduce assumptions about the age pattern of mortality, which assumption should be as accurate as possible.; The age pattern of mortality drawn directly from the intercensal life table of the population under study is hypothesized to be the best candidate. To make an intercensal life table for China's population, three methods are discussed and compared with each other. They are Preston's method, Coale's method and a new method developed in this study. The new method is found to be the best in terms of the least error produced.; Then an estimated age pattern of the intercensal mortality is taken as a general standard, is combined with the Brass logit system improved by adding a characteristic term to the system, and is used to solve for annual survival ratio functions of life tables by iteration.; Based on the estimated survival ratio functions, annual life tables and annual age distributions of the population are constructed from 1953 to 1982. The completeness of birth and death registration is then assessed. The excess number of deaths during the 1959-1961 famine period is also estimated.