R: functions to determine distance of multivariate data to normal distribution

Question

I have a multivariate data and I am interested to compute the distance of complete data to multivariate normal distribution. I want to use R. I have seen some functions like shapiro-wilk test etc. But from them I can only understand if p-value is less <0.05 it does not follow normal distribution. But I want to know how much it is far from the normal distribution. Can anyone please refer me to some functions that I can refer to for use.

Answer 1

Use the mqqnorm function from the RVAideMemoire package. It shows, among others, Mahalanobis distances. From the function example:

x <- 1:30+rnorm(30)
y <- 1:30+rnorm(30,1,3)
mqqnorm(cbind(x,y))