Modernizing Demographic Computations: Some Illustrative Examples

Abstract

Although the use of computers and sophisticated software, particularly R, has become common in demography, this has not led to a revision of some traditional demographic algorithms, which were originally designed to overcome computational obstacles that nowadays no longer exist. This article argues that, rather than implementing the existing algorithms in today’s powerful software, it is time to take advantage of this software to modify the algorithms themselves. It illustrates this idea with three examples: Sprague interpolation, the Relational Gompertz method for fertility estimation, and the estimation of infant and child mortality from proportions of surviving children by mother’s age. In each case, it is shown that the analysis can be made simpler, more direct,intuitive, transparent, and precise by making use of the resources offered by software such as R. It is also shown that the greater flexibility of these algorithms creates opportunities for extending each method.