Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Multiple-Testing Approach to the Multivariate Behrens-Fisher ProblemOn Testing Equality of Covariance Matrices

A Multiple-Testing Approach to the Multivariate Behrens-Fisher Problem: On Testing Equality of... [In this chapter, we present two approaches for testing equality of covariance matrices. In the complete-data case, Box’s M method is presented. The Type I errors and power of Box’s M method are presented. In the randomly-incomplete-data case, a new method is proposed. This method uses the False Discovery Rate (FDR) algorithm of Benjamini and Hochberg (J. R. Stat. Soc. Series B. 57, 1289–1300, 1995). The Type I errors and power in the randomly-incomplete-case are also presented. An example is also provided.] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

A Multiple-Testing Approach to the Multivariate Behrens-Fisher ProblemOn Testing Equality of Covariance Matrices

Part of the SpringerBriefs in Statistics Book Series
Desai, Tejas
Springer Journals — Jan 29, 2013
Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/a-multiple-testing-approach-to-the-multivariate-behrens-fisher-problem-0YQ80601hh
Publisher
Springer New York
Copyright
© The Author 2013
ISBN
978-1-4614-6442-6
Pages
17 –29
DOI
10.1007/978-1-4614-6443-3_3
Publisher site
See Chapter on Publisher Site

Abstract

[In this chapter, we present two approaches for testing equality of covariance matrices. In the complete-data case, Box’s M method is presented. The Type I errors and power of Box’s M method are presented. In the randomly-incomplete-data case, a new method is proposed. This method uses the False Discovery Rate (FDR) algorithm of Benjamini and Hochberg (J. R. Stat. Soc. Series B. 57, 1289–1300, 1995). The Type I errors and power in the randomly-incomplete-case are also presented. An example is also provided.]

Published: Jan 29, 2013

Keywords: Alternative hypothesis; Box’s M statistic; Covariance matrices; False discovery rate; Null hypothesis; Power; Type I error

There are no references for this article.