AD.test {mvnTest}

Anderson-Darling test for multivariate normality

Description

This function implements the Anderson-Darling test for assessing multivariate normality. It calculates the value of the test and its approximate p-value.

Usage

 AD.test(data, qqplot = FALSE)

Arguments

data

A numeric matrix or data frame.

qqplot

If TRUE produces a chi-squared QQ plot.

Value

AD

the value of the test statistic.

p.value

the p-value of the test.

Note

The printing method and plotting are in part adapted from R package MVN (version 4.0, Korkmaz, S. et al., 2015).

The computations are relatively expensive as Monte Carlo procedure is used to calculate empirical p-vales.

Author(s)

Rashid Makarov, Vassilly Voinov, Natalya Pya

References

Paulson, A., Roohan, P., and Sullo, P. (1987). Some empirical distribution function tests for multivariate normality. Journal of Statistical Computation and Simulation, 28, 15-30

Henze, N. and Zirkler, B. (1990). A class of invariant consistent tests for multivariate normality. Communications in Statistics - Theory and Methods, 19, 3595-3617

Selcuk Korkmaz, Dincer Goksuluk, and Gokmen Zararsiz. MVN: Multivariate Normality Tests, 2015. R package version 4.0

See Also

S2.test, CM.test, DH.test, R.test, HZ.test

Examples

## Not run: 
## generating n bivariate normal random variables...       
dat <- rmvnorm(n=100,mean=rep(0,2),sigma=matrix(c(4,2,2,4),2,2)) 
res <- AD.test(dat)
res

## generating n bivariate t distributed with 10df random variables...   
dat <- rmvt(n=200,sigma=matrix(c(4,2,2,4),2,2),df=10,delta=rep(0,2)) 
res1 <- AD.test(dat)
res1

data(iris)
setosa <- iris[1:50, 1:4] # Iris data only for setosa
res2 <- AD.test(setosa, qqplot = TRUE)
res2

    
## End(Not run)