Shrinking Large-Scale Population Projection Models by Aggregation and Decomposition

Rogers, A. (1975). Shrinking Large-Scale Population Projection Models by Aggregation and Decomposition. IIASA Working Paper. IIASA, Laxenburg, Austria: WP-75-167

Preview

Text WP-75-167.pdf Download (1MB) | Preview

Abstract

During the past two decades social scientists have come to model dynamic socioeconomic systems of growing size and complexity. Despite a heavy reliance on ever more sophisticated high-speed digital computers, however, their capacity for handling such systems has not kept pace with the growing demands for more detailed information.

An increasing number of social scientists currently find themselves in the somewhat frustrating position of being asked to provide accurate projections at very fine levels of detail with resources that are scarcely sufficient for carrying out such projections at much more aggregate levels of resolution. Prominent among them are demographers who are called upon to produce consistent projections of regional populations disaggregated by age, color, race, sex, and such indicators of class and welfare as employment category and income. Since the computational requirements of this task are staggering, the need for developing improved methods for "shrinking" population projection models by reducing their dimensionality is an urgent one, and the two most obvious methods for effecting such a reduction are aggregation and partitioning, or more appropriately, decomposition.

Actions (login required)

View Item