Biometrical Letters Vol. 51(2), 2014, pp. 103-124
|
ANALYSIS OF MULTIVARIATE REPEATED MEASURES DATA USING A MANOVA MODEL AND PRINCIPAL COMPONENTS Mirosław Krzysko1, Tadeusz Śmiałowski2, Waldemar Wołynski1 1Faculty of Mathematics and Computer Science, Adam Mickiewicz University, Umultowska 87, 61-614 Poznan, Poland, e-mail: mkrzysko@amu.edu.pl, wolynski@amu.edu.pl 2Plant Breeding and Acclimatization Institute, National Research Institute, Radzików, 05-870 Błonie, Poland, e-mail: zhsmialo@cyf-kr.edu.pl |
In this paper we consider a set of T repeated measurements on p characteristics on each of n individuals. The n individuals themselves may be divided and randomly assigned to K groups. These data are analyzed using a mixed effect MANOVA model, assuming that the data on an individual have a covariance matrix which is a Kronecker product of two positive definite matrices. Results are illustrated on a data set obtained from experiments with varieties of winter rye.
multivariate repeated measures data (doubly multivariate data), Kronecker product covariance structure, maximum likelihood estimates, mixed MANOVA model, principal component analysis
