Multivariate analysis of variance: Difference between revisions

Revision as of 06:31, 26 December 2018

Revision as of 07:00, 11 February 2019

Line 21:

|last=Chiani | first=M.

|year=2016

|title = Distribution of the largest root of a matrix for Roy’sRoy's test in multivariate analysis of variance

|journal=[[Journal of Multivariate Analysis]]

|volume=143

|pages=467–471

|arxiv=1401.3987v3

| doi=10.1016/j.jmva.2015.10.007

}}</ref> while the distribution under the alternative is studied in.<ref>I.M. Johnstone, B. Nadler "Roy's largest root test under rank-one alternatives" arXiv preprint arXiv:1310.6581 (2013)</ref>

The best-known [[approximation]] for Wilks' lambda was derived by [[C. R. Rao]].

Line 33 ⟶ 34:

==Correlation of dependent variables==

MANOVA's power is affected by the correlations of the dependent variables and by the effect sizes associated with those variables. For example, when there are two groups and two dependent variables, MANOVA's power is lowest when the correlation equals the ratio of the smaller to the larger standardized effect size.<ref>{{cite journal|last1=Frane|first1=Andrew|title=Power and Type I Error Control for Univariate Comparisons in Multivariate Two-Group Designs|journal=Multivariate Behavioral Research|volume=50|issue=2|pages=233–247|date=2015|doi=10.1080/00273171.2014.968836|pmid=26609880}}</ref>

==See also==