Theorem 12. The eigenvalues of a Hermitian matrix [or symmetric real matrix] are real. The eigenvalues of a skew-Hermitian matrix [or skew-symmetric real matrix] are zero or pure imaginary. The eigenvalues of a unitary [or real orthogonal matrix] all have absolute value equal to 1.
Theorem 13. The eigenvectors belonging to different eigenvalues of a Hermitian matrix [or symmetric real matrix] are orthogonal.
Theorem 14. [Cayley-Hamilton] A matrix satisfies its own characteristic equation.
Theorem 15. [reduction of matrix to diagonal form] If a non-singular matrix
has distinct eigenvalues
with corresponding eigenvectors written as columns in the matrix
![]()
then
![]()
i.e.
, called the transform of
by
, is a diagonal matrix containing the eigenvalues of
in the main diagonal and zeros elsewhere. We say that
has been transformed or reduced to diagonal form.
Theorem 16. [Reduction of quadratic form to canonical form] Let
be a symmetric matrix, for example,
![]()
Then if
, we obtain the quadratic form
![]()
The cross product terms of this quadratic form can be removed by letting
where
is the column vector with elements
and
is an orthogonal matrix which diagonalizes
. The new quadratic form in
with no cross product terms is called the canonical form. A generalization can be made to Hermitian quadratic forms.
Theorems on eigenvalues and eigenvectors
