Hi there. Firstly thanks for a great resource!

I am trying to decide which is the best method for performing PCA on a matrix X, of dimension m x n. As far as i understand it there are two options:

1) Find eigenvectors (and vals) of covariance matrix

2) Perform SVD on matrix Y, where Y = 1/sqrt(n-1)X^T

Any help/advice would be much appreciated as i am fairly new to this.
Cheers!

SVD is almost always the method of choice for PCA.
In fact, PCA basically *is* SVD.