Algorithmic Approximation of Optimal Value Differential Stability Bounds in Nonlinear Programming,
Abstract
The data needed to calculate the sensitivity to data perturbations of the solution and optimal value of a mathematical program are available as by-products of many solution. Fiacco demonstrated this is developing a penalty function technque for approximating the parameter derivatives of a solution for quite general perturbed non-linear programs. Armacost and Mylander used this to advantage in making available the routine calculation of sensitivity information as part of a computer code for the Sequential Unconstrained Minimization Technique (SUMT).
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1981
- Accession Number
- ADA110545
Entities
People
- William P. Hutzler
Organizations
- RAND Corporation