## Principal Components

For principal components from a covariance matrix,
the names of the variables containing principal
component scores are **PCV1**, **PCV2**, **PCV3**, and so on.
The output component scores are a linear combination
of the centered **Y** variables with coefficients equal
to the eigenvectors of the covariance matrix.
For principal components from a correlation matrix,
the names of the variables containing principal
component scores are **PCR1**, **PCR2**, **PCR3**, and so on.
The output component scores are a linear combination
of the standardized **Y** variables with coefficients
equal to the eigenvectors of the correlation matrix.

If you specify **Variance=Eigenvalues** in the multivariate method
options dialog, the new variables of principal component scores have
mean zero and variance equal to the associated eigenvalues. If you
specify **Variance=1**, the new variables have variance equal to
one.

Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.