Hi,
"In PCA, the first principal direction is the eigenvector of the data matrix with the largest eigenvalue" is stated false. Could you please explain why ?
It's of the matrix X_T*X not the data matrix X.
Because it's the eigenvector of matrix X^T * X. Sometimes there are traps in the wording of the answers, we have to be careful ;)
But they have the same matrix U and V so why isn't it equivalent ?
PCA - Q6 2018
Hi,
"In PCA, the first principal direction is the eigenvector of the data matrix with the largest eigenvalue" is stated false.
Could you please explain why ?
It's of the matrix X_T*X not the data matrix X.
3
Because it's the eigenvector of matrix X^T * X.
Sometimes there are traps in the wording of the answers, we have to be careful ;)
But they have the same matrix U and V so why isn't it equivalent ?
Add comment