The function takes a design-with-responses object created by xy_Obj
and produces 50-50 MANOVA output. Results are produced for each term in the
model.
manova5050(xyObj, stand)design-with-responses object
standardisation of responses (0 or 1)
A list with components
model term names
explained variances calculated from sums of squares summed over all responses
degrees of freedom - adjusted for other terms in model
degrees of freedom - adjusted for terms contained in actual term
number of principal components used for testing
number of principal components used as buffer components
variance explained by nPC components
variance explained by (nPC+nBU) components
50-50 MANOVA p-values
logical. Whether the responses are standardised.
Classical multivariate ANOVA (MANOVA) are useless in many practical cases. The tests perform poorly in cases with several highly correlated responses and the method collapses when the number of responses exceeds the number of observations. 50-50 MANOVA is made to handle this problem. Principal component analysis (PCA) is an important part of this methodology. Each test is based on a separate PCA.
Langsrud, <U+00D8>. (2002) Rotation Tests. The Statistician, 51, 305--317.