Munich Personal RePEc Archive

A SAS® Macro for the Generalized RAS Algorithm

Coleman, Charles (2016): A SAS® Macro for the Generalized RAS Algorithm.

WarningThere is a more recent version of this item available.

Download (4MB) | Preview


Demographers and economists frequently encounter the problem of constraining matrices of mixed sign to controls of possibly mixed sign. The recently developed Generalized RAS (GRAS) algorithm is presented to solve these problems. The GRAS algorithm produces a unique solution that minimizes an entropy-like function. The algorithm is applied to a well-known example and compared to the solution originally obtained using a generalization of the Akers-Siegel procedure.

Available Versions of this Item

MPRA is a RePEc service hosted by
the Munich University Library in Germany.