Show simple item record

dc.contributor.authorBouffier, Amanda Marie
dc.date.accessioned2014-03-04T21:05:14Z
dc.date.available2014-03-04T21:05:14Z
dc.date.issued2013-08
dc.identifier.otherbouffier_amanda_m_201308_ms
dc.identifier.urihttp://purl.galileo.usg.edu/uga_etd/bouffier_amanda_m_201308_ms
dc.identifier.urihttp://hdl.handle.net/10724/28981
dc.description.abstractThe Maximally Informative Next Experiment (MINE) criterion was developed for designing large, expensive genomics experiments. Four variations of the MINE method for the linear model were created: MINE-like, MINE, MINE with random orthonormal basis, and MINE with rotation for the linear model. Theorem 1 establishes sufficient conditions for the maximization of a MINE criterion under the linear model. Theorem 2 is established when the MINE criterion is equivalent to the classic design criterion of D-optimality. By simulation under the linear model, we establish that the MINE with random orthonormal basis and MINE with random rotation are faster to discover the true linear relation with p regression coefficients and n observations when p >> n. These two variations also display a lower false positive rate than MINE or MINE-like methods for a least a majority of the experiments.
dc.languageeng
dc.publisheruga
dc.rightspublic
dc.subjectLinear Model
dc.subjectMaximally Informative Next Experiment (MINE)
dc.subjectModel-Guided Discovery
dc.subjectD-optimality
dc.titleA MINE alternative to D-optimal designs for the linear model
dc.typeThesis
dc.description.degreeMS
dc.description.departmentBioinformatics
dc.description.majorBioinformatics
dc.description.advisorJonathan Arnold
dc.description.committeeJonathan Arnold
dc.description.committeeThiab Taha
dc.description.committeeBernd Schuttler


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