Show simple item record

dc.contributor.authorLiu, Xinyu
dc.date.accessioned2014-03-04T21:13:19Z
dc.date.available2014-03-04T21:13:19Z
dc.date.issued2013-08
dc.identifier.otherliu_xinyu_201308_phd
dc.identifier.urihttp://purl.galileo.usg.edu/uga_etd/liu_xinyu_201308_phd
dc.identifier.urihttp://hdl.handle.net/10724/29106
dc.description.abstractSingle-nucleotide polymorphisms (SNPs), believed to determine human differences, are widely used to predict risk of diseases and class membership of subjects. In the literature, several supervised machine learning methods, such as, support vector machine, neural network and logistic regression, are available for classification. Typically, however, samples for training a machine are limited and/or the sampling cost is high. Thus, it is essential to determine the minimum sample size needed to construct a classifier based on SNP data. Such a classifier would facilitate correct classification while keeping the sample size to a minimum, thereby making the studies cost-effective. In this dissertation, for coded SNP data from two classes, an optimal classifier and an approximation to its probability of correct classification (PCC) are derived. A linear classifier is constructed and an approximation to its PCC is also derived. These approximations are validated through a variety of Monte Carlo simulations. A sample size determination algorithm based on the criterion which ensures that the difference between the two approximate PCCs is below a threshold, is given. For the HapMap data on Chinese and Japanese populations, a linear classifier is built using 51 independent SNPs, and the required total sample sizes are determined using our algorithm. For coded SNP data from D(>=2) classes, we derive an optimal Bayes classifier and a linear classifier, and obtain a normal approximation to the PCC for each classifier. These approximations are used to evaluate the associated Area Under the Receiver Operating Characteristic (ROC) Curve (AUCs) or Volume Under the ROC hyper-Surface (VUSs). We give an algorithm for sample size determination, which ensures that the difference between the two approximate AUCs (or VUSs) is below a pre-specified threshold. The performance of this algorithm is also illustrated via simulations. For the HapMap data with three and four populations, a linear classifier is built using 92 independent SNPs and the required total sample sizes are determined. We also illustrate the usefulness of our sample size determination algorithm in a prediction problem using a Heterogeneous Stock Mice data.
dc.languageeng
dc.publisheruga
dc.rightspublic
dc.subjectArea Under the Receiver Operating Characteristic Curve
dc.subjectClassification
dc.subjectHapmap data
dc.subjectHeterogeneous Stock Mice data
dc.subjectProbability of correct classification
dc.subjectReceiver Operating Characteristic
dc.subjectSample Size Determination
dc.subjectSingle-nucleotide polymorphisms
dc.subjectVolume Under the Receiver Operating Characteristic hyper-Surface
dc.subjectWald test.
dc.titleSample size determination in multi-class classification and prediction based on single-nucleotide polymorphisms
dc.typeDissertation
dc.description.degreePhD
dc.description.departmentStatistics
dc.description.majorStatistics
dc.description.advisorT. N. Sriram
dc.description.committeeT. N. Sriram
dc.description.committeeXiangrong Yin
dc.description.committeeLily Wang
dc.description.committeeJaxk Reeves
dc.description.committeeWilliam P. McCormick


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record