The decision on dimensionality of the space spanned by general linear
functions of the parameter matrix of a MANOVA model is considered. Thi
s problem is related to the investigation, whether graphically or anal
ytically, of significant empirical departures From the overall null hy
pothesis on these functions. A closed testing procedure for a sequence
of relevant hypotheses is proposed. Unlike the classical procedures b
ased on asymptotic distributions of the likelihood ratio statistics, t
he proposed method ensures that the Type I familywise error rate does
not exceed the nominal cc-level. Also, it is consistent with testing t
he overall null hypothesis, while relying on tests of subsequent linea
r hypotheses implied by the former. Examples are given to compare the
proposed procedure with a classical one. (C) 1998 Academic Press.