Real spectral theorem
Statement and proof
\(\mc{M}^{n \times n}\) denotes the set of \(n \times n\) matrices. \(\mc{S}^{n \times n} \subset \mc{M}^{n \times n}\) denotes the set of symmetric real valued \(n \times n\) matrices.
Theorem 1 Real spectral theorem
Let \(\bsf{F} \in \mc{S}^{n \times n}\). Then there exists an orthonormal basis of \(\mbb{R}^{n}\) consisting of the eigenvectors of \(\bsf{F}\) and a diagonal matrix \(\bsf{D}\) such that \[ \bsf{F} = \bsf{N} \bsf{D} \bsf{N}^{\msf{T}} \] where \(\bsf{N} \in \mc{M}^{n \times n}\) whose columns are the eigenvectors of \(\bsf{F}\) and the elements of the diagonal matrix \(\bsf{D}\) are the eigenvalues of \(\bsf{F}\). Further, the matrices \(\bsf{N}\) and \(\bsf{D}\) need not be unique.
Example 1 Let \(n = 3\). Then \(\bsf{D}\) and \(\bsf{N}\) can be written as \[ \bsf{D} = \begin{pmatrix} \lambda_{1} & 0 & 0 \\ 0 & \lambda_{2} & 0 \\ 0 & 0 & \lambda_{3} \end{pmatrix} \quad \bsf{N} = \begin{pmatrix} ^{1}N_{1} & ^{2}N_{1} & ^{3}N_{1} \\ ^{1}N_{2} & ^{2}N_{2} & ^{3}N_{2} \\ ^{1}N_{3} & ^{2}N_{3} & ^{3}N_{3} \end{pmatrix} \] where \(\lambda_{i}\) is the \(i\)-th eigenvalue (the eigenvalues need not be ordered) and \(^{d}N_{i}\) is the \(i\)-th component of the \(d\)-th eigenvector.
Proof 1. Proof of the real spectral theorem
To be completed.
The spectral theorem also implies that the symmetric matrix \(\bsf{F}\) can be written as \[ \bsf{F} = \sum_{i = 1}^{n} \lambda_{i} \, ^{i}\bsf{N} \otimes ^{i}\bsf{N} \]
So, if \(n = 3\) for example, \[ \bsf{F} = \lambda_{1} \, ^{1}\bsf{N} \otimes ^{1}\bsf{N} + \lambda_{2} \, ^{2}\bsf{N} \otimes ^{2}\bsf{N} + \lambda_{3} \, ^{3}\bsf{N} \otimes ^{3}\bsf{N} \]