Numerical tests for multivariate normality
Besides the graphical tests using QQ-plot, the MVN
library has a range of numerical tests for checking multivariate normality. Along with the significance tests, these tests provide an inferential statement about whether the data are multivariate normal.
This exercise is part of the course
Multivariate Probability Distributions in R
Exercise instructions
Use the
mvn()
function with argumentmvnTest = "mardia"
to perform Mardia's test of multivariate normality on themultnorm.sample
dataset.Use the
mvn()
function with argumentmvnTest = "hz"
to perform Henze-Zirkler's test of multivariate normality on the numeric variables,Alcohol
,Malic
,Ash
,Alcalinity
of thewine
dataset.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Mardia's test
# Henze-Zirkler's test