Advertisement

Simultaneous Inferences for Ratio Parameters Using Multiple Contrasts Test

  • Dan LinEmail author
  • Gemechis D. Djira
  • Ziv Shkedy
  • Tomasz Burzykowski
  • Ludwig A. Hothorn
Chapter
Part of the Use R! book series (USE R)

Abstract

In Chap. 16, we wish to search for genes for which the mean gene-expression increases by 100 ×δ% from the control for at least one (perhaps the highest) dose, where δ is a relative biological ratio of interest. Genes can be tested for a pre-specified δ and can be ranked based on the significance of the ratio test. Genes found significant with a large δ are of higher interest to the biologists. We discuss the ratio test statistic and the link between Williams- and Marcus-type multiple contrasts tests (MCTs) and the corresponding ratio testing procedures

References

  1. Bretz, F. (2006). An extension of the Williams trend test to general unbalanced linear models. Computational Statistics & Data Analysis, 50, 1735–1748.Google Scholar
  2. Dilba, D., Bretz, F., Guiard, V., & Hothorn, L. A. (2004). Simultaneous confidence intervals for ratios with application to the comparison of several treatments with a control. Method of Information in Medicine 43, 465–469.Google Scholar
  3. Dilba, G., Bretz, F., Hothorn, L. A., & Guiard, V. (2006a). Power and sample size computations in simultaneous tests for non-inferiority based on relative margins. Statistics in Medicine, 25, 1131–1147.Google Scholar
  4. Dilba, G., Brets, F., & Guiard, V. (2006b). Simultaneous confidence sets and confidence intervals for multiple ratios. Journal of Statistical Planning and Inference, 136, 2640–2658.Google Scholar
  5. Djira, G. D. (2010). Relative potency estimation in parallel-line assys—Method comparison and some extensions. Communications in Statistics—Theory and Methods, 39, 1180–1189.Google Scholar
  6. Dilba, G, Schaarschmidt, F., & Hothorn, L. A. (2007). Inferences for ratios of normal means. RNews 7, 1, 20–23.Google Scholar
  7. Hothorn, L. A., & Djira, G. D. (2010). Ratio-to-control Williams-type test for trend. Pharmaceutical Statistics. doi:10.1002/pst.464.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Dan Lin
    • 1
    Email author
  • Gemechis D. Djira
    • 3
  • Ziv Shkedy
    • 2
  • Tomasz Burzykowski
    • 2
  • Ludwig A. Hothorn
    • 4
  1. 1.Veterinary Medicine Research and DevelopmentPfizer Animal HealthZaventemBelgium
  2. 2.Interuniversity Institute for Biostatistics and Statistical Bioinformatics (I-BioStat), Center for Statistics (CenStat)Hasselt UniversityDiepenbeekBelgium
  3. 3.Department of Mathematics and StatisticsSouth Dakota State UniversityBrookingsUSA
  4. 4.Institute of BiostatisticsLeibniz University HannoverHannoverGermany

Personalised recommendations