Get a sorted matrix of eigen vectors

I have to write a function that takes in X (N*D data matrix) and return eigenvecs a D*D matrix, normalized eigenvectors (with length=1) sorted based on its eigenvalue magnitude (with d-th column corresponds to eigenvector with the d-th biggest eigenvalue).

How can I use eig to solve this problem. eig take sonly square matrix but X here id N*D

Answer 1

[Tushar Aggarwal]

Normalize X to have 0 mean

X=X-mean(X(:));

calculate covariance

covx = cov(X);

and then eignvectors and eigenvalue

 [a,b] = eig(covx);