Eigenvalues in principal component analysis

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Posts: 2
Joined: Tue Apr 03, 2012 11:37 pm

Eigenvalues in principal component analysis

Postby stat_student » Tue Apr 03, 2012 11:54 pm

Hi all,

need your help.

I run principal component analysis on the data, and get specific eigenvalues in the SPSS output (in the table "Total variance explained" -> "Initial eigenvalues"). However, when I compute eigenvalues "manually", they differ from those computed with PCA.

Particularly, I use the following syntax commands for "manual" computation in SPSS:

compute xtx=transpos(x)*x.
call eigen(xtx,eigenvectors,eigenvalues)
print eigenvalues.

where x is the matrix that contains my variables in the columns. Then I compute x transposed * x, and compute eigenvalues of this matrix.

Interestingly, eigenvalues are different from those I get with PCA.

What can be the problem? Does PCA method in SPSS compute eigenvalues in a different way?

Thanks in advance for any help.


Who is online

Users browsing this forum: No registered users and 1 guest