Despite the apparent discrepancy, the scalar increment-decrement model of Schoen (1975), described by equations (3) and (4), is consistent with the Rogers-Ledent matrix formulation shown in equation (I). Either approach may be preferable in certain specific instances. In general, however, the greater power of the matrix formulation and its more explicit recognition of both age and region give it substantial advantages over the scalar model.

The text of this article is only available as a PDF.
You do not currently have access to this content.