5. Multiple Tests

  1. Phillip I. Good

Published Online: 23 FEB 2011

DOI: 10.1002/9780470937273.ch5

Analyzing the Large Number of Variables in Biomedical and Satellite Imagery

Analyzing the Large Number of Variables in Biomedical and Satellite Imagery

How to Cite

Good, P. I. (2011) Multiple Tests, in Analyzing the Large Number of Variables in Biomedical and Satellite Imagery, John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9780470937273.ch5

Author Information

  1. Huntington Beach CA, USA

Publication History

  1. Published Online: 23 FEB 2011
  2. Published Print: 21 MAR 2011

ISBN Information

Print ISBN: 9780470927144

Online ISBN: 9780470937273

SEARCH

Keywords:

  • false discovery rate (FDR);
  • heuristic methods;
  • multivariate procedure;
  • overall error rate;
  • univariate statistics

Summary

This chapter considers a variety of ways to control either the overall error or the false discovery rate (FDR). The reduction proceeds in stages, first identifying which genes are differentially expressed and then selecting those of the greatest significance. Selection methods fall into two broad categories: computation of univariate statistics and rank, and heuristic methods. The list of univariate statistics that have been employed to make pairwise comparisons is a lengthy one and includes the t-test, the Wilcoxon, empirical Bayes t-statistic, Welch t-statistic, fold change, and area under the receiver operating characteristic curve. Several studies have examined feature selection by investigating the consistency between gene lists from small subsets of samples and those from the full data set or using a bootstrap method to generate simulated datasets from real data sets. The chapter also discusses a multivariate procedure specific to the analysis of microarrays.

Controlled Vocabulary Terms

Bayesian statistics; bootstrap method; false discovery rate; multivariate statistics; one-sample wilcoxon signed rank test