Modal matrix

In linear algebra, the modal matrix is used in the diagonalization process involving eigenvalues and eigenvectors.^[1]

Specifically the modal matrix ${\displaystyle M}$ for the matrix ${\displaystyle A}$ is the n × n matrix formed with the eigenvectors of ${\displaystyle A}$ as columns in ${\displaystyle M}$. It is utilized in the similarity transformation

${\displaystyle D=M^{-1}AM,}$

where ${\displaystyle D}$ is an n × n diagonal matrix with the eigenvalues of ${\displaystyle A}$ on the main diagonal of ${\displaystyle D}$ and zeros elsewhere. The matrix ${\displaystyle D}$ is called the spectral matrix for ${\displaystyle A}$. The eigenvalues must appear left to right, top to bottom in the same order as their corresponding eigenvectors are arranged left to right in ${\displaystyle M}$.^[2]

The matrix

${\displaystyle A={\begin{pmatrix}3&2&0\\2&0&0\\1&0&2\end{pmatrix}}}$

has eigenvalues and corresponding eigenvectors

${\displaystyle \lambda _{1}=-1,\quad \,\mathbf {b} _{1}=\left(-3,6,1\right),}$

${\displaystyle \lambda _{2}=2,\qquad \mathbf {b} _{2}=\left(0,0,1\right),}$

${\displaystyle \lambda _{3}=4,\qquad \mathbf {b} _{3}=\left(2,1,1\right).}$

A diagonal matrix ${\displaystyle D}$, similar to ${\displaystyle A}$ is

${\displaystyle D={\begin{pmatrix}-1&0&0\\0&2&0\\0&0&4\end{pmatrix}}.}$

One possible choice for an invertible matrix ${\displaystyle M}$ such that ${\displaystyle D=M^{-1}AM,}$ is

${\displaystyle M={\begin{pmatrix}-3&0&2\\6&0&1\\1&1&1\end{pmatrix}}.}$^[3]

Note that since eigenvectors themselves are not unique, and since the columns of both ${\displaystyle M}$ and ${\displaystyle D}$ may be interchanged, it follows that both ${\displaystyle M}$ and ${\displaystyle D}$ are not unique.^[4]

Let ${\displaystyle A}$ be an n × n matrix. A generalized modal matrix ${\displaystyle M}$ for ${\displaystyle A}$ is an n × n matrix whose columns, considered as vectors, form a canonical basis for ${\displaystyle A}$ and appear in ${\displaystyle M}$ according to the following rules:

• All Jordan chains consisting of one vector (that is, one vector in length) appear in the first columns of ${\displaystyle M}$.
• All vectors of one chain appear together in adjacent columns of ${\displaystyle M}$.
• Each chain appears in ${\displaystyle M}$ in order of increasing rank (that is, the generalized eigenvector of rank 1 appears before the generalized eigenvector of rank 2 of the same chain, which appears before the generalized eigenvector of rank 3 of the same chain, etc.).^[5]

One can show that

${\displaystyle AM=MJ,}$

1

where ${\displaystyle J}$ is a matrix in Jordan normal form. By premultiplying by ${\displaystyle M^{-1}}$, we obtain

${\displaystyle J=M^{-1}AM.}$

2

Note that when computing these matrices, equation (1) is the easiest of the two equations to verify, since it does not require inverting a matrix.^[6]

This example illustrates a generalized modal matrix with four Jordan chains. Unfortunately, it is a little difficult to construct an interesting example of low order.^[7] The matrix

${\displaystyle A={\begin{pmatrix}-1&0&-1&1&1&3&0\\0&1&0&0&0&0&0\\2&1&2&-1&-1&-6&0\\-2&0&-1&2&1&3&0\\0&0&0&0&1&0&0\\0&0&0&0&0&1&0\\-1&-1&0&1&2&4&1\end{pmatrix}}}$

has a single eigenvalue ${\displaystyle \lambda _{1}=1}$ with algebraic multiplicity ${\displaystyle \mu _{1}=7}$. A canonical basis for ${\displaystyle A}$ will consist of one linearly independent generalized eigenvector of rank 3 (generalized eigenvector rank; see generalized eigenvector), two of rank 2 and four of rank 1; or equivalently, one chain of three vectors ${\displaystyle \left\{\mathbf {x} _{3},\mathbf {x} _{2},\mathbf {x} _{1}\right\}}$, one chain of two vectors ${\displaystyle \left\{\mathbf {y} _{2},\mathbf {y} _{1}\right\}}$, and two chains of one vector ${\displaystyle \left\{\mathbf {z} _{1}\right\}}$, ${\displaystyle \left\{\mathbf {w} _{1}\right\}}$.

An "almost diagonal" matrix ${\displaystyle J}$ in Jordan normal form, similar to ${\displaystyle A}$ is obtained as follows:

${\displaystyle M={\begin{pmatrix}\mathbf {z} _{1}&\mathbf {w} _{1}&\mathbf {x} _{1}&\mathbf {x} _{2}&\mathbf {x} _{3}&\mathbf {y} _{1}&\mathbf {y} _{2}\end{pmatrix}}={\begin{pmatrix}0&1&-1&0&0&-2&1\\0&3&0&0&1&0&0\\-1&1&1&1&0&2&0\\-2&0&-1&0&0&-2&0\\1&0&0&0&0&0&0\\0&1&0&0&0&0&0\\0&0&0&-1&0&-1&0\end{pmatrix}},}$

${\displaystyle J={\begin{pmatrix}1&0&0&0&0&0&0\\0&1&0&0&0&0&0\\0&0&1&1&0&0&0\\0&0&0&1&1&0&0\\0&0&0&0&1&0&0\\0&0&0&0&0&1&1\\0&0&0&0&0&0&1\end{pmatrix}},}$

where ${\displaystyle M}$ is a generalized modal matrix for ${\displaystyle A}$, the columns of ${\displaystyle M}$ are a canonical basis for ${\displaystyle A}$, and ${\displaystyle AM=MJ}$.^[8] Note that since generalized eigenvectors themselves are not unique, and since some of the columns of both ${\displaystyle M}$ and ${\displaystyle J}$ may be interchanged, it follows that both ${\displaystyle M}$ and ${\displaystyle J}$ are not unique.^[9]

• Beauregard, Raymond A.; Fraleigh, John B. (1973), A First Course In Linear Algebra: with Optional Introduction to Groups, Rings, and Fields, Boston: Houghton Mifflin Co., ISBN 0-395-14017-X
• Bronson, Richard (1970), Matrix Methods: An Introduction, New York: Academic Press, LCCN 70097490
• Nering, Evar D. (1970), Linear Algebra and Matrix Theory (2nd ed.), New York: Wiley, LCCN 76091646