The NLP Procedure |
The parameter vector may be subject to a set of linear equality and inequality constraints:
The linear constraints define a feasible region in that must contain the point that minimizes the problem. If the feasible region is empty, no solution to the optimization problem exists.
All optimization techniques in PROC NLP (except those processing nonlinear constraints) are active set methods. The iteration starts with a feasible point , which either is provided by the user or can be computed by the Schittkowski and Stoer (1979) algorithm implemented in PROC NLP. The algorithm then moves from one feasible point to a better feasible point along a feasible search direction :
Theoretically, the path of points never leaves the feasible region of the optimization problem, but it can hit its boundaries. The active set of point is defined as the index set of all linear equality constraints and those inequality constraints that are satisfied at . If no constraint is active for , the point is located in the interior of , and the active set is empty. If the point in iteration hits the boundary of inequality constraint , this constraint becomes active and is added to . Each equality or active inequality constraint reduces the dimension (degrees of freedom) of the optimization problem.
In practice, the active constraints can be satisfied only with finite precision. The LCEPSILON= option specifies the range for active and violated linear constraints. If the point satisfies the condition
If you cannot expect an improvement in the value of the objective function by moving from an active constraint back into the interior of the feasible region, you use this inequality constraint as an equality constraint in the next iteration. That means the active set still contains the constraint . Otherwise you release the active inequality constraint and increase the dimension of the optimization problem in the next iteration.
A serious numerical problem can arise when some of the active constraints become (nearly) linearly dependent. Linearly dependent equality constraints are removed before entering the optimization. You can use the LCSINGULAR= option to specify a criterion used in the update of the QR decomposition that decides whether an active constraint is linearly dependent relative to a set of other active constraints.
If the final parameter set is subjected to linear equality or active inequality constraints, the QR decomposition of the matrix of the linear constraints is computed by , where is an orthogonal matrix and is an upper triangular matrix. The columns of matrix can be separated into two matrices, , where contains the first orthogonal columns of and contains the last orthogonal columns of . The column-orthogonal matrix is also called the nullspace matrix of the active linear constraints . The columns of the matrix form a basis orthogonal to the rows of the matrix .
At the end of the iteration process, the PROC NLP can display the projected gradient
Those elements of the vector of first-order estimates of Lagrange multipliers
Copyright © 2008 by SAS Institute Inc., Cary, NC, USA. All rights reserved.