The assumption of multivariate normality is often at the basis of many statistical analysis. In this article we propose a graphical method based on the characterization of the multivariate normal distribution in terms of univariate normality of all linear combinations of the variables in the set. We review some methods for choosing directions to look for departure from the hypothesis of normality, and we propose an interactive dynamic graphic approach for checking the joint distribution. Examples from simulated data and a real dataset will be used to show practical implementation of the ideas outlined.
Assessing multivariate normality through interactive dynamic graphics
SCRUCCA, Luca
2000
Abstract
The assumption of multivariate normality is often at the basis of many statistical analysis. In this article we propose a graphical method based on the characterization of the multivariate normal distribution in terms of univariate normality of all linear combinations of the variables in the set. We review some methods for choosing directions to look for departure from the hypothesis of normality, and we propose an interactive dynamic graphic approach for checking the joint distribution. Examples from simulated data and a real dataset will be used to show practical implementation of the ideas outlined.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
