On the relationship between regression analysis and mathematicalprogramming

摘要

The interaction between linear, quadratic programming and regression analysisare explored by both statistical and operations research methods. Estimation andoptimization problems are formulated in two different ways: on one hand linear andquadratic programming problems are formulated and solved by statistical methods, andon the other hand the solution of the linear regression model with constraints makesuse of the simplex methods of linear or quadratic programming. Examples are given toillustrate the ideas.