File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Block-diagonal discriminant analysis and its bias-corrected rules

TitleBlock-diagonal discriminant analysis and its bias-corrected rules
Authors
Issue Date2013
Citation
Statistical Applications in Genetics and Molecular Biology, 2013, v. 12 n. 3, p. 347-359 How to Cite?
AbstractHigh-throughput expression profiling allows simultaneous measure of tens of thousands of genes at once. These data have motivated the development of reliable biomarkers for disease subtypes identification and diagnosis. Many methods have been developed in the literature for analyzing these data, such as diagonal discriminant analysis, support vector machines, and k-nearest neighbor methods. The diagonal discriminant methods have been shown to perform well for high-dimensional data with small sample sizes. Despite its popularity, the independence assumption is unlikely to be true in practice. Recently, a gene module based linear discriminant analysis strategy has been proposed by utilizing the correlation among genes in discriminant analysis. However, the approach can be underpowered when the samples of the two classes are unbalanced. In this paper, we propose to correct the biases in the discriminant scores of blockdiagonal discriminant analysis. In simulation studies, our proposed method outperforms other approaches in various settings. We also illustrate our proposed discriminant analysis method for analyzing microarray data studies. © 2013 Walter de Gruyter GmbH, Berlin/Boston.
Persistent Identifierhttp://hdl.handle.net/10722/194520
ISSN
2011 Impact Factor: 1.517
2015 SCImago Journal Rankings: 0.954
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorPang, H-
dc.contributor.authorTong, T-
dc.contributor.authorNg, M-
dc.date.accessioned2014-01-30T03:32:41Z-
dc.date.available2014-01-30T03:32:41Z-
dc.date.issued2013-
dc.identifier.citationStatistical Applications in Genetics and Molecular Biology, 2013, v. 12 n. 3, p. 347-359-
dc.identifier.issn1544-6115-
dc.identifier.urihttp://hdl.handle.net/10722/194520-
dc.description.abstractHigh-throughput expression profiling allows simultaneous measure of tens of thousands of genes at once. These data have motivated the development of reliable biomarkers for disease subtypes identification and diagnosis. Many methods have been developed in the literature for analyzing these data, such as diagonal discriminant analysis, support vector machines, and k-nearest neighbor methods. The diagonal discriminant methods have been shown to perform well for high-dimensional data with small sample sizes. Despite its popularity, the independence assumption is unlikely to be true in practice. Recently, a gene module based linear discriminant analysis strategy has been proposed by utilizing the correlation among genes in discriminant analysis. However, the approach can be underpowered when the samples of the two classes are unbalanced. In this paper, we propose to correct the biases in the discriminant scores of blockdiagonal discriminant analysis. In simulation studies, our proposed method outperforms other approaches in various settings. We also illustrate our proposed discriminant analysis method for analyzing microarray data studies. © 2013 Walter de Gruyter GmbH, Berlin/Boston.-
dc.languageeng-
dc.relation.ispartofStatistical Applications in Genetics and Molecular Biology-
dc.titleBlock-diagonal discriminant analysis and its bias-corrected rules-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1515/sagmb-2012-0017-
dc.identifier.pmid23735433-
dc.identifier.scopuseid_2-s2.0-84881622680-
dc.identifier.volume12-
dc.identifier.issue3-
dc.identifier.spage347-
dc.identifier.epage359-
dc.identifier.isiWOS:000321247600004-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats