Gábor Szűcs, Faculty of mathematics, physics and informatics, Comenius University in Bratislava, Slovak Republic

Pages: 20 – 33

Abstract

Analysis of variance was primarily designed to assess similarities and differences between more groups and by means of which the effect of one or more factors can be tested. If various statistical features are being pursued simultaneously on the observed objects, it is recommended to use the multivariate analysis of variance offering more complex possibilities of the difference testing between two or more groups or effects. This paper provides a brief introduction to the theory of onedimensional analysis of variance and the generalization of the model to the multivariate case. It presents the theoretical apparatus for hypothesis testing using the multivariate analysis of variance and its illustrations on real data sets in R software.