STATISTICAL CONTROL FOR NON-LINEAR OPTIMIZATION.
Abstract
The paper delineates an algorithm which provides a statistical point estimate as well as a statistical upper confidence point for the global optimum, say g*, in a non-convex mathematical programming problem. Further, the procedure will supply a value of the input vector, x dot, so that g(x dot), the objective function evaluated at x dot, is close to g*. Two examples involving the optimization of second order response surface designs with known global optima are given for the purpose of comparing the point estimate and the upper confidence point estimate with the known values. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1969
- Accession Number
- AD0693507
Entities
People
- Herman Otto Hartley
- R. C. Pfaffenberger
Organizations
- Texas A&M University