When does a square matrix have an eigen-decomposition? When is a matrix defective?

3.6k Views Asked by At

Some square matrices, like $ \left( \begin{array}{cc} 0 & 1 \\ 0 & 0 \end{array} \right)$, don't have a complete set of eigenvectors. By complete I mean that the eigenvectors span the entire space. The only eigenvector of the previous matrix is $\left( \begin{array}{c} 1 \\ 0 \end{array} \right)$. These are called defective matrices. How can one prove whether a matrix is or is not defective?

The reason that I ask is that I know stochastic matrices are supposed to have an eigen-decomposition but I don't know how to show it?

1

There are 1 best solutions below

0
On

A matrix is not defective if and only if is diagonalizable, so if you calculate the characteristic polynomial of a matrix $A$, that is $det(A-tI)$, you can find the algebraic multiplicity of any eigenvalue. Now you must calculate the geometric multiplicity of any eigenvalue $\lambda$ that is the size of $Ker(A-\lambda I)$ and if it is equal to the geometric multiplicity for any eigen value $\lambda$ you have that the matrix is diagonalizable and is not defective.