Providing software solutions since 1976
Operations Research and Project Management Papers
Nonlinear Optimization in SAS/OR® Software: Migrating from PROC NLP to PROC OPTMODEL
Tao Huang and Edward P. Hughes, SAS Institute, 2010
Abstract
PROC OPTMODEL, the flagship optimization procedure in SAS/OR® software, is also intended to supersede PROC NLP in the long run for nonlinear optimization. The rich and flexible PROC OPTMODEL syntax enables natural and compact algebraic formulations of nonlinear models. In addition to supporting the major legacy algorithms present in PROC NLP, PROC OPTMODEL has access to powerful new algorithms that can solve large-scale problems and handle nonlinearity more robustly. These new algorithms exploit sparse structures and implement the state-of-the-art techniques in numerical optimization.
In addition to benefiting from ease of modeling and access to improved algorithms, PROC OPTMODEL users can use its programming capabilities to develop customized solution methods. PROC OPTMODEL has many features that make it an excellent replacement for PROC NLP not only for operations research practitioners but also for statisticians, econometricians, and data miners. This paper uses several examples to illustrate how to migrate from PROC NLP to PROC OPTMODEL and highlight the benefits of doing so.
