It is shown that in the MANOVA situation, certain dependent tests of hypothesis may be combined to obtain an overall conservative significance level. The result holds whether one uses the Wilks, Roy, Lawley-Hotelling or Pillai test criterion. This generalizes a result due to Kimball in univariate ANOVA.
@article{1176344628,
author = {Dykstra, Richard L.},
title = {On Dependent Tests of Significance in the Multivariate Analysis of Variance},
journal = {Ann. Statist.},
volume = {7},
number = {1},
year = {1979},
pages = { 459-461},
language = {en},
url = {http://dml.mathdoc.fr/item/1176344628}
}
Dykstra, Richard L. On Dependent Tests of Significance in the Multivariate Analysis of Variance. Ann. Statist., Tome 7 (1979) no. 1, pp. 459-461. http://gdmltest.u-ga.fr/item/1176344628/