Principal component analysis provides the information needed to understand the role of the original descriptors in the formation of the principal components. It may also be used to show the relationships among original descriptors in the reduced space. The role of descriptors in principal component analysis is examined in this Subsection under various aspects: matrix of eigenvectors, projection in reduced space (matrix
UA1/2), and projection in reduced space (matrix U).
1. The matrix of eigenvectors — In Subsection 1 (above), the relationships among the normalized eigenvectors (which are the columns of the square matrix U) were studied using an expression of the form U'U. For the numerical example:
Was this article helpful?