Mskekur

Mardia's multivariate skewness and kurtosis coefficients and its hypotheses testing.

Calculates the Mardia's multivariate skewness and kurtosis coefficients as well as their corresponding statistical tests. For large sample size the multivariate skewness is asymptotically distributed as a Chi-square random variable; here it is corrected for small sample size. Likewise, the multivariate kurtosis it is distributed as a unit-normal.

Inputs:
X - multivariate data matrix [Size of matrix must be n(data)-by-p (variables)].
c - normalizes covariance matrix by n (c=1[default]) or by n-1 (c~=1)
alpha - significance level (default = 0.05).

Outputs:
-Complete statistical analysis table of both multivariate Mardia's skewness and kurtosis.
-Chi-square quantile-quantile (Q-Q) plot of the squared Mahalanobis distances of the observations from the mean vector.
-The file ask you whether or not are you interested to label the n data points on the Q-Q plot:
Are you interested to explore all the n data points? (y/n):