In PCA, the PC's are eigenvectors of the the covariance (or correlation) matrix \\(R\\) of a given p-by-n data matrix \\(X\\).

When I load the example, I can view the model and use my mouse to rotate and drag it. However when I start typing a random key ...