Term of Award

Fall 2004

Degree Name

Master of Applied Mathematics

Document Type and Release Option

Thesis (restricted to Georgia Southern)

Department

Department of Mathematical Sciences

Committee Chair

Goran Lesaja

Committee Member 1

Richard Hathaway

Committee Member 2

Yingkang Hu

Abstract

The stratified sampling problem is a very important problem that arises in different applications and very often in statistics The purpose of this thesis is to numerically test two interior point methods on the stratified sampling problem. First, a homogeneous conic quadratic interior point method was used to solve the stratified sampling problem as a reformulated conic quadratic problem Then, a nonlinear homogeneous interior point method was used to solve the stratified sampling problem as a reformulated monotone complementarity problem Finally, the methods were compared by CPU time and the number of iterations showing that the first approach works better for problems with higher dimensions.

Files over 10MB may be slow to open. For best results, right-click and select "Save as..."

Share

COinS