Could someone help to explain to me why this statement is wrong? In PCA, the first principal direction is the eigenvector of the data matrix X with the largest associated eigenvalue. Thanks in advance.
The eigenvectors are of the covariance matrix XX.T or X.TX, not X alone.
I think this is correct: In PCA, the first principal direction is the eigenvector of the data matrix XX.T with the largest associated eigenvalue.
It's actually the singular values, not eigenvalues.
I think both singular values and eigenvalues are correct, since eigenvalues of matrix XX.T are squared singular values of data matrix X.
PCA in the exam 2019
Could someone help to explain to me why this statement is wrong?
In PCA, the first principal direction is the eigenvector of the data matrix X with the largest associated eigenvalue.
Thanks in advance.
The eigenvectors are of the covariance matrix XX.T or X.TX, not X alone.
I think this is correct:
In PCA, the first principal direction is the eigenvector of the data matrix XX.T with the largest associated eigenvalue.
2
It's actually the singular values, not eigenvalues.
1
I think both singular values and eigenvalues are correct, since eigenvalues of matrix XX.T are squared singular values of data matrix X.
1
Add comment