MAHAL return the Mahalanobis' D-square distance between the 
 multivariate samples x and y, which must have the same number 
 of components (columns), but may have a different number of observations (rows). 
 
  d = mahal(X,Y)

   d(k) = (X(k,:)-MU)*inv(SIGMA)*(X(k,:)-MU)'

   where MU and SIGMA are the mean and the covariance matrix of Y 


 see also: TRAIN_SC, TEST_SC, COVM

 References: 

Package: nan