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