DEVELOPMENT OF STATISTICAL DISCRIMINANT MATHEMATICAL-PROGRAMMING MODEL VIA RESAMPLING ESTIMATION TECHNIQUES

This paper uses resampling estimation techniques to develop a statisti cal mathematical programming model for discriminant analysis problems. Deleted-d jackknife, deleted-d bootstrap, and bootstrap procedures ar e used to identify statistical significant parameter estimates for a d iscriminant mathematical programming (MP) model. The results of this p aper indicate that the resampling approach is a viable model selection technique. Furthermore, estimating the MP models via resampling techn iques can also improve the classification performance compared to a de terministic discriminant MP model. In this study, the deleted-d jackkn ife procedure was the most promising among the resampling estimation t echniques examined.