The Effect of Certain Modifications to Mathematical Programming Models for the Two-Group Classification Problem Page: 4
The following text was automatically extracted from the image on this page using optical character recognition software:
estimated on both training samples and validation (holdout)
samples. Exact misclassification rates are determined from
the estimated classification functions for some models.
Several factors, such as sample size, covariance structure,
distribution, and orientation of the data, are varied in the
The results show that the modified mathematical
programming models have potential for being very useful in
situations in which violations of the usual parametric
assumptions are severe. This study addresses certain issues
in implementing mathematical programming approaches to the
classification problem. For example, with some mathematical
programming models, there are solutions that are not
invariant under data translations or rotations. The study
shows the usefulness of a general contaminated multivariate
normal distribution in estimating misclassification
probabilities. The study also illustrates that a wide range
of values can be assigned to the measures of skewness and
kurtosis when generating the contaminated normal
distribution by using different parameter settings. The
results of this study will assist practitioners in
understanding and implementing improved versions of
mathematical programming formulations and, thus, give them
greater flexibility in choosing an appropriate
Here’s what’s next.
This dissertation can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Dissertation.
Wanarat, Pradit. The Effect of Certain Modifications to Mathematical Programming Models for the Two-Group Classification Problem, dissertation, May 1994; Denton, Texas. (digital.library.unt.edu/ark:/67531/metadc278866/m1/4/: accessed February 23, 2019), University of North Texas Libraries, Digital Library, digital.library.unt.edu; .