Population projections are ordinarily carried out by applying to an initial age distribution a set of rates of ferti1ity and survivorship which moves the population along through a time period, say five years; then applying to the resultant age distri...
Population projections are ordinarily carried out by applying to an initial age distribution a set of rates of ferti1ity and survivorship which moves the population along through a time period, say five years; then applying to the resultant age distribution the same or a modified set of fertility and survivorship rates, and so continuing as long as the results appear to be of interest. The disentangling of the projection from the age distribution to which it is applied, and its separate analysis, are easily carried out by viewing the projection as a matrix. It will be seen, among other arithmetical results, that the dominant latent root of the population matrix is the exponential of the intrinsic rate of natural increase.
We proceed to reduce the matrix for five-year groups to much smaller dimensions, attempting to find approximations that will retain its essential features. We will confine the following discussion to our assumptions of high mortality, no immigration and no war losses for native-born Korean fmeales The first reduction in the arithmetical work is to confine the calculation to females, and to that portion of them which has not passed beyond reproductive age. One may think of women and girls up to the age of 45 as the dynamic portion of any self-contained human group.
In this paper we estimate Korean population from 1975 to. 2000 by a matrix, operator which is developed by N. Keyfitz.